Functional and Structural Analysis of the Metal-Ion Transporter ZIP14

zhao_n.pdf

TRP

ORF

PBS

PNGase F

qPCR

UTR

WT

Zip/ZIP

Zrt


Transient receptor potential cation channel

Open reading frame

Phosphate-buffered saline

Peptide N-glycosidase F

Quantitative polymerase chain reaction

Untranslated region

Wild-type

Zrt- and Irt-like proteins

Zinc-regulated transporter









deficiency (atransferrinemia and hypotransferrinemia) (51,126,127). Affected individuals

absorb dietary iron efficiently and deposit large amounts in the liver even though they

lack TF. It has been shown that after TF saturation, 70% of the NTBI given orally or

intravenously, was taken by the liver (128). The molecular basis of NTBI uptake is not

well understood.

In summary, the liver can take up both TBI and NTBI. DMT1, the only known iron

transporter involved in TBI release from endosomes, may also mediate NTBI into cells.

However, DMT1 knockout mice can still accumulate iron in the liver, indicating that

DMT1 is not the major iron transporter in either pathway for this organ. How iron is

taken up into the liver, especially into hepatocytes, remains to be elucidated (9).

ZIP14 and Iron Metabolism

Identification and Characterization of ZIP14

ZIP14 (SLC39A14, solute carrier 39 family, member 14), as one of the members of

ZIP metal-ion transporter superfamily, can transport both zinc and iron (129). It is the

second identified iron-import protein to date (the first one is DMT1). The ZIP family

protein takes the name from ZRT, IRT like protein, where ZRT represents zinc-

regulated transporter and IRT stands for iron-regulated transporter (130).

By sequencing clones obtained from the cDNA library of a human immature

myeloid cell line (KG-1 cells), Nomura et al. (131) first cloned ZIP14, which they named

KIAA0062. Northern blot analysis detected that ZIP14 was ubiquitously expressed, with

highest expression in the liver, and lowest expression was in the spleen, thymus, and

peripheral blood leukocytes. A multiple tissue expression array also showed ubiquitous

expression of ZIP14 with high expression in the liver, pancreas and heart (132). By

searching databases for sequences that were similar to a unique HEXPHEXGD motif (a









I' P<0.05

500-

400-

300-

E 200-
E
o 100

0
Con Zipl 4


B

Control Zip 4

anti-Zipl4
-- 50 KDa

anti-TFR1
-- 75 KDa

anti-DMT1


- 60 KDa
- 50 KDa


anti-Actin


Figure 3-1. Overexpression of Zipl4 increases TBI uptake. A) HEK 293T cells
transfected with pCMVSport2 (control) or pCMVSport6/mouse Zipl4 were
incubated with 100 nM 59Fe-TF for 4 h. Cells were harvested and cell-
associated radioactivity was determined by y-counting. The results are
representative of one of three independent experiments without significant
variation between experiments. B) Lysates of cells incubated with 100 nM
59Fe-TF for 4 h were analyzed by Western blotting for Zipl4, TFR1 and
DMT1. To indicate protein loading among lanes, blot was stripped and
reprobed for actin. Data are representative of one of three experiments.


I~









Future Directions

This study was performed exclusively in cell culture models. Future studies using

whole animals will be required to define the in vivo role of ZIP14 in iron metabolism,

especially in the liver. Utilization of knockout mouse models has been powerful for

revealing the function of genes in vivo. Generation of a global and tissue-specific Zip 4

gene knockout mouse will be needed to demonstrate the role of Zipl4 in iron biology.

Comparative genomics has shown that the mouse and human genomes share

high degree of homology, suggesting the mouse also serves as a model for finding new

therapeutic interventions for human diseases (208). Iron disorders, such as

hemochromatosis, is one of the genetic diseases attracting more attention from

researchers. The C282Y mutation in HFE gene leads to the most common form of

hereditary hemochromatosis (209). Since the discovery of the HFE gene in 1996 (140),

much effort has been devoted to identify its precise function. In HH patients, excess iron

deposits in the liver, mainly in hepatocytes. Neither the import nor the export pathway

for iron in hepatocytes is well understood (9). Hfe gene knockout (Hfe'-) mice displayed

liver iron-loading phenotype, similar to human HH disease. The only known iron

importer DMT1 was considered to account for the iron deposition, but survival rate of

DMT1'- mice was greater when inactivating Hfe together. Hfe'-DMT1-'- animal continued

to deposit iron in the liver during growth (86). These data suggest that iron transporter

other than DMT1 is involved in hepatic iron loading throughout the development of HH.

It has been found overexpression of HFE led to decreased ZIP14 stability and reduced

both TBI and NTBI uptakes in HepG2 cells, suggesting the interaction of these two

proteins. Further research is needed to test this hypothesis by using gene knockout









BIOGRAPHICAL SKETCH

Ningning Zhao was born in Jinhua, Zhejiang province, China. He received a

Bachelor of Science degree and a Master of Education degree from Beijing Sport

University, Beijing, China. He majored in exercise biochemistry and focused on sports

nutrition. He did a thesis research about iron and exercise-induced anemia, which

interested him to pursue further knowledge in iron metabolism.

In June 2006, he came to the Department of Food Science and Human Nutrition at

University of Florida for the PhD program in nutritional sciences. He worked with Dr.

Mitchell Knutson and focused on molecular aspects of iron metabolism, which

broadened his understanding about iron biology and inspired him to continue learning in

the iron field.

After graduation, he will go to Oregon Health and Science University for a

postdoctoral training in iron biology at the laboratory of Dr. Caroline Enns.









endogenous ZIP14 is also glycosylated as the ZIP14 band shifted down with PNGase F

treatment (Fig.4-4C). This demonstrates that Zipl4 is an N-linked glycoprotein.

Identification of N-linked Glycosylation Sites in mZipl4

To identify which predicted asparagines are indeed glycosylated in mZipl4, I

mutated the five potential sites by replacing asparagines (N) with aspartic acid (D)

singly or in combination. The effect of these mutations were examined by transiently

expressing wide-type (WT) and mutants in HEK 293T cells, followed by Western

analysis of the collected cell lysates. I observed that in the HEK 293T overexpression

system, WT-mZipl4 always exhibited two bands around 50 kDa. Single replacement of

asparagines at positions 52, 75, 85 and 100 all led to a decrease in the molecular mass

of the upper band, suggesting that each site is glycosylated, whereas mutation of

asparagines 455 did not affect the band pattern compared to WT. After removal of N-

glycosylation sites in tandem, the upper band showed a stepwise decrease as well.

With deletion of all four asparagines sites, only the lower band was detected (Fig.4-5A).

The same result was observed by using the C-T-Flag mZipl4 wide-type and mutant

constructs (Fig.4-5B). These results indicate that the first four predicted asparagines

near the amino- terminus of mZipl4 are indeed glycosylated, whereas asparagine 455

is not.

Schematic Membrane Topology Model of mZipl4

Based on the epitope mapping results, the identified glycosylation sites, and

multiple computer program predictions (Table 4-1), A membrane topology model of

mZipl4 protein was proposed (Fig. 4-6). In this model, mZipl4 has seven

transmembrane regions with an extracellular amino- terminus and an intracellular

carboxy- terminus. The histidine-rich region is also located intracellularly.









Nonpermeabilized Permeabilized
Qnnn- n I CQnnnin


EEA1





B


N-T- Flag
mZip14




C


His Flag
mZipl4




D C-T- Flag
mZipl4








Figure 4-2. Flag epitope mapping of mZipl4. HEK 293T cells were transiently
transfected with pCMVSport2 (control) or Flag-tagged constructs A) EEA1
can only be detected under permeabilized condition, indicating the antibody
can pass through cell membrane only after saponin treatment. B) N-T-Flag-
mZipl4 can be detected by Flag antibody under both permeabilized and
nonpermeabilized conditions. C-D) His-Flag-mZipl4 and C-T-Flag-mZipl4
can only be detected by Flag antibody under permeabilized condition. Data
are representative of three independent experiments.









females have less iron stores due to periodic blood loss through menstruation (4). In a

balanced state, approximately 0.5 to 2 mg of dietary iron is absorbed through duodenal

enterocytes every day, and the same amount is lost in the urine, feces, sweat and

sloughed cells (5,6). About 65% of the iron in the body is incorporated into hemoglobin

of red blood cells (RBCs) and 10% is present in myoglobin, other enzymes and

cytochromes. The remaining body iron is stored in the liver, macrophages of the

reticuloendothelial system (RES) and bone marrow (4,7,8).

Iron Homeostasis

Regulation of systemic iron homeostasis involves intricate control of intestinal iron

absorption, effective erythropoietic iron utilization, efficient iron recycling from effete

erythrocytes, and controlled iron storage by hepatocytes and macrophages (9).

Intestinal Iron Absorption

Since deficiency of iron results in anemia, while overload leads to formation of

reactive oxygen species (ROS), iron homeostasis needs to be tightly controlled. The

systemic regulation of iron homeostasis takes place primarily in intestinal iron

absorption because humans have no physiologic pathway to excrete excess iron (4).

Dietary iron is found in two basic forms, either as heme, found in meat and meat

products or non-heme iron, present in vegetables, beans, and fruits. Non-heme iron

predominates in all diets, comprising 90%-95% of total daily iron intake. The regulation

of iron absorption relies on mechanisms that sense dietary iron content as well as iron

storage levels in the body and erythropoietic iron requirements (10). The absorption of

iron is illustrated in Figure 1-2. Dietary iron uptake occurs at the apical membrane of

duodenal enterocytes. The insoluble ferric form of iron is the primary non-heme iron in

food and must be reduced to ferrous iron before transporting across the intestinal










Dietary iron
Duodenum
1-2 mg/day



Iron usage
Muscle
(Myoglobin) Circulating iron
300 mg
300 mg Plasma diferric TF
3mg Storage iron
r? i/Liver
S(Parenchymal cells)
1000 mg





Bone marrow
(Erythroblasts)
300 mg Spleen
) (Macrophages)
600 mg


Red blood cells
(Hemoglobin)
1800 mg





Iron loss
Sloughed mucosal cells
Desquamation
Menstruation
Other blood loss
S1-2 mg per day




Figure 1-1. Distribution of iron within the body. Normal adults typically have 3-5 g of total
body iron. To maintain iron balance, about 1-2 mg of dietary iron is absorbed
every day to replace the iron that is lost in the urine, feces, sweat and
sloughed cells. Most body iron can be found in mature erythrocytes and in
erythroid bone marrow. Reticuloendothelial macrophages recycle iron from
old red cells and supply most of the iron for new red blood cell synthesis.
Approximately 0.1% of body iron is bound to TF, a circulating plasma protein
that delivers iron to erythroid precursors and other tissues. Iron is stored
primarily in hepatocytes.









TBI uptake without decreasing TF uptake in HepG2 cells. These results suggest that

ZIP14 participates in the uptake of iron from TF, thus identifying a potentially new role

for ZIP14 in iron metabolism.

Most ZIP proteins have been predicted to contain eight transmembrane (TM)

helices with both amino- (N-) and carboxy- (C-) termini located extracellularly. Structural

analysis by Flag epitope mapping and bioinformatic prediction support a model of

mZip14 that has seven TM segments. It was also found that the N- terminus is an

ectodomain, whereas the C-terminus and the long loop containing a histidine-rich,

putative metal-binding motif localize intracellularly. In addition, N-linked glycosylation

sites were identified through site-directed mutagenesis, and it was found that mZip14 is

glycosylated at asparagines 52, 75, 85 and 100, residues that are all in the extracellular

N- terminus. It was also demonstrated that N-linked glycosylation is not required for cell-

surface localization but it is required for iron transport activity.

In conclusion, a new role of ZIP14 in iron metabolism was identified-ZIP14

mediates the uptake of iron bound to TF. I also investigated the membrane topology of

mZip14 and conclude that it has seven transmembrane regions. Finally, I found that N-

linked glycosylation is essential for the iron transport ability of Zip14, but not for its

trafficking to the plasma membrane.









marrow rely primarily on the endosomal TF-TFR1 pathway for iron uptake. Rapid

development of liver iron overload in these animals indicates that insufficient utilization

of iron in erythrocytes triggers the increased absorption of iron from intestine. The

identical mutation in DMT1 of b rats and mk mice, together with three reported cases of

DMT1 mutations in humans, emphasizes the role of DMT1 in erythropoiesis (94). In

addition, DMT1 is expressed in bone marrow and optimally functions at acidic pH, such

as in endosomes (18). These observations support the rationale that DMT1 is an

endosomal iron transport protein, especially in erythroid cells.

Subcellular Localization of DMT1

Studies of the subcellular localization of DMT1 have been done mainly by

categorizing DMT1 into the +IRE isoform and the -IRE isoform. DMT1(+IRE) protein is

expressed in the duodenum where it is regulated by dietary iron (83). DMT1(-IRE) is

expressed in erythroid cell precursors where it is regulated by erythropoietin (95).

Besides their tissue-specific expression and regulation, these two isoforms may function

in different subcellular compartments. In CHO (Chinese hamster ovary) cells which

stably expressed the DMT1(-IRE) form, a ring-like staining at the periphery of the cells

was observed, indicating plasma membrane localization. When cells were

permeabilized, an intracellular punctate pattern could be seen in addition to membrane

staining, indicating its localization in intracellular compartments (96). In CHO cells, RAW

cells (mouse leukemic monocyte macrophage cell line), MEL cells (mouse

erythroleukemia cell line), and TM4 cells (mouse sertoli cell line), DMT1 displayed

plasma membrane staining and clear colocalization with TF in recycling endosomes

(97).









147. Krogh, A., Larsson, B., von Heijne, G., and Sonnhammer, E. L. (2001) J. Mol.
Biol. 305, 567-580

148. Tusnady, G. E., and Simon, I. (2001) Bioinformatics (Oxford, England) 17, 849-
850

149. Claros, M. G., and von Heijne, G. (1994) Comput. Appl. Biosci. 10, 685-686

150. Cao, B., Porollo, A., Adamczak, R., Jarrell, M., and Meller, J. (2006)
Bioinformatics (Oxford, England) 22, 303-309

151. Hofmann, K., and Stoffel, W. (1993) Biol. Chem. Hoppe-Seyler 347, 166

152. Arai, M., Mitsuke, H., Ikeda, M., Xia, J. X., Kikuchi, T., Satake, M., and Shimizu,
T. (2004) Nucleic Acids Res. 32, W390-393

153. Jones, D. T., Taylor, W. R., and Thornton, J. M. (1994) Biochemistry 33, 3038-
3049

154. Kall, L., Krogh, A., and Sonnhammer, E. L. (2007) Nucleic Acids Res. 35, W429-
432

155. Hirokawa, T., Boon-Chieng, S., and Mitaku, S. (1998) Bioinformatics (Oxford,
England) 14, 378-379

156. Bendtsen, J. D., Nielsen, H., von Heijne, G., and Brunak, S. (2004) J. Mol. Biol.
340, 783-795

157. Fariselli, P., Finocchiaro, G., and Casadio, R. (2003) Bioinformatics (Oxford,
England) 19, 2498-2499

158. Hiller, K., Grote, A., Scheer, M., Munch, R., and Jahn, D. (2004) Nucleic Acids
Res. 32, W375-379

159. Hansen, J. E., Lund, O., Tolstrup, N., Gooley, A. A., Williams, K. L., and Brunak,
S. (1998) Glycoconj. J. 15, 115-130

160. Graham, R. M., Morgan, E. H., and Baker, E. (1998) Eur. J. Biochem. 253, 139-
145

161. Knutson, M. D. (2007) Nutr. Rev. 65, 335-340

162. Morgan, E. H., Smith, G. D., and Peters, T. J. (1986) Biochem. J. 237, 163-173

163. Scheiber, B., and Goldenberg, H. (1996) Arch. Biochem. Biophys. 326, 185-192


113









LIST OF REFERENCES ........... .... .............................. ............. .............. 104

B IO G RA PH ICA L S KETC H ......................................................... ............................. 117


















































7












N-T-3xFlag
mZipl 4


anti-Flag
(Zip14)


B
C-T-3xFlag 0,
mZipl4 _-


anti-Flag
(Zipl4)


S- 50 KDa


... ..... .- 5 0K.D a

*~ WW^~. 50 KDa


Figure 4-5. Identification of N-linked glycosylation sites in mZipl4. A) Western blot
analysis of cell lysates from HEK 293T cells transiently transfected with empty
vector (pCMVSport2, Sport2) or N-T-Flag mZipl4 expression vectors. Zipl4
expression constructs encode either wild-type (WT) or N-glycosylation
site/sites mutants (N52D, N75D, N85D, N100D, N455D, N52/75D,
N52/75/85D and N52/75/85/100D). Position 52, 75, 85, and 100 are identified
to be glycosylated, but position 455 is not. B) Western analysis by using C-T-
Flag mZipl4 WT construct and mutants also indicates that the first four
asparagines are linked with sugar and the last one is not. In both A and B, the
samples were electrophoresed on 10% polyacrylamide gel, transferred to
nitrocellulose, and probed with anti-Flag antibody for Zipl4.









ZIP14 participates in the uptake of iron from TF, thus identifying a potentially new role

for ZIP14 in iron metabolism.

The second part of my project was to study the structure of ZIP14 protein. I used

mZipl4 to study its membrane topology and glycosylation effects. To investigate the

membrane topology of mZipl4, a Flag epitope was inserted into the N-terminus, C-

terminus, as well as the long extramembrane domain containing a histidine-rich metal-

binding motif. The tagged proteins were expressed in HEK 293T cells, and the

accessibility of the Flag tags by antibody was determined by immunofluorescence

analysis of intact and permeabilized cells. Based on the experimental results together

with bioinformatic predictions, It was concluded that mZipl4 has seven transmembrane

domains, with an extracellular N-terminus, an intracellular C-terminus and a cytoplasmic

large loop which contains histidine-rich metal-binding motif. Furthermore, glycosylation

sites were identified by mutating each of the 5 potential N-linked glycosylation sites. The

mutants were transiently expressed in HEK 293T cells, followed by Western blotting. I

found that mZipl4 is glycosylated at asparagines 52, 75, 85 and 100, residues that are

all in the extracellular amino- terminus, confirming mZipl4 is an N-linked glycoprotein.

Lastly, to examine the role of glycosylation in plasma membrane trafficking and iron

transport activity, mZipl4 wide-type and asparagine mutant (which has no N-linked

glycosylation sites) were transiently expressed in HEK 293T cells. I found that N-

glycosylation of mZipl4 is not required for cell-surface localization but it is required for

iron transport activity.


100









164. Trinder, D., and Morgan, E. (1997) Hepatology 26, 691-698

165. McArdle, H. J., Danzeisen, R., Fosset, C., and Gambling, L. (2003) Biometals 16,
161-167

166. Kralova, A., Svetlikova, M., Madar, J., Ulcova-Gallova, Z., Bukovsky, A., and
Peknicova, J. (2008) Reprod. Biol. Endocrinol. 6, 27

167. Trinder, D., Olynyk, J. K., Sly, W. S., and Morgan, E. H. (2002) Proc. Natl. Acad.
Sci. U. S. A. 99, 5622-5626

168. Kawabata, H., Nakamaki, T., Ikonomi, P., Smith, R. D., Germain, R. S., and
Koeffler, H. P. (2001) Blood98, 2714-2719

169. Wetli, H. A., Buckett, P. D., and Wessling-Resnick, M. (2006) Chem. Biol. 13,
965-972

170. Zhang, A. S., Canonne-Hergaux, F., Gruenheid, S., Gros, P., and Ponka, P.
(2008) Exp. Hematol. 36, 1227-1235

171. van Renswoude, J., Bridges, K. R., Harford, J. B., and Klausner, R. D. (1982)
Proc. Natl. Acad. Sci. U. S. A. 79, 6186-6190

172. Bokhari, M., Carnachan, R. J., Cameron, N. R., and Przyborski, S. A. (2007) J.
Anat. 211, 567-576

173. Ikuta, K., Zak, 0., and Aisen, P. (2004) Int. J. Biochem. Cell Biol. 36, 340-352

174. Morgan, E. H., and Baker, E. (1986) Fed. Proc. 45, 2810-2816

175. Trinder, D., Morgan, E. H., and Baker, E. (1988) Biochim Biophys Acta 943, 440-
446

176. Herbison, C. E., Thorstensen, K., Chua, A. C., Graham, R. M., Leedman, P.,
Olynyk, J. K., and Trinder, D. (2009) Am. J. Physiol. Cell Physiol. 297, C1567-
1575

177. Graham, R. M., Chua, A. C., Herbison, C. E., Olynyk, J. K., and Trinder, D.
(2007) World J. Gastroenterol. 13, 4725-4736

178. Eide, D. J. (2004) Pflugers Arch. 447, 796-800

179. Gaither, L. A., and Eide, D. J. (2000) J. Biol. Chem. 275, 5560-5564

180. Gaither, L. A., and Eide, D. J. (2001) J. Biol. Chem. 276, 22258-22264


114









P>0.05


P<0.01

I


20000


. 15000
0)
05

10000
E
E
0 5000


P<0.01
F


Sport2 WT mZipl4 ASNs mutant


Sport2 WT


anti-Flag
(Zip14)



anti-Tubulin


ASNs
M i itont


- 50 KDa
- 50 KDa


Figure 4-8. Functional analysis of mZipl4 lacking N-linked glycosylation. A) The iron
transport activity of wild-type and mutant mZipl4 was analyzed by measuring
the NTBI uptake 48 h after transient expression in HEK 293T cells. After three
times washing and incubation with SFM for 1 h, cells transfected with empty
pCMVSport2 (Sport2), N-T-Flag mZipl4 (WT mZipl4) or N-T-Flag mZipl4
asparagines mutant which has disruption in all four glycosylated asparagines
(ASNs mutant) were incubated with 2 pM 59Fe-ferric citrate for 2 h in uptake
buffer at 37C. The amount of 59Fe taken up by cells is expressed as cpm
per mg of protein. Data represent the mean SEM of three independent
experiments. B) Cell lysates harvested 48 h after transient transfection were
analyzed by immunoblot for Zipl4 and tubulin as a lane loading control.




































Figure 4-6. Schematic illustration of mZipl4 membrane topology. Mouse Zipl4 has 7
transmembrane segments with an extracellular N-terminus and an
intracellular C-terminus. Histidine-rich region, also intracellular, is indicated by
individual white circles of histidines (H). Black circles of asparagine (N)
indicate N-linked glycosylation sites.









rapid internalization of TF-TFR1 complex upon binding of the holo-TF (diferric TF) to its

receptor, with formation of a coated pit and a coated vesicle which sheds its clathrin to

become an endosome. Once internalized, endosomes are acidified by an ATP-

dependent proton pump (46). Acidification weakens binding of ferric iron to TF and

causes conformational changes of both TF and TFR1 (47). Dissociated ferric iron is

reduced to the ferrous form by Steap3 (six-transmembrane epithelial antigen of the

prostate 3), the dominant ferrireductase in the erythroid TF cycle (14,48). After

reduction, Fe2+ is transported into the cytosol by DMT1. Both TF and TFR1 will return to

the cell surface, where TF is released for further usage. The TF cycle is depicted in

Figure 1-4.

Iron Utilization by Erythrocytes

Erythrocytes require iron for the oxygen-carrying capacity of hemoglobin. The

erythroid bone marrow is the largest consumer of iron. Failure to incorporate adequate

iron into heme results in impaired erythrocyte maturation and leads to microcytic

hypochromic anemia. In healthy individuals, about two-thirds of the total body iron is

accounted for by hemoglobin in developing erythroid precursors and mature RBCs.

Approximately 20 25 mg of iron are needed every day for hemoglobin synthesis (26).

Nearly all circulating iron is bound to TF in the plasma. TF carries one or two atoms of

iron per protein molecule. Erythroid precursors meet their iron needs by taking up TBI

through the TF cycle (26). The amount of TFR1 present on the cell surface determines

the amount of iron imported into cells. The expression of TFR1 is regulated

developmentally during erythoid maturation, correlating with the changing rates of

hemoglobin production (49).









proximal duodenum (18,83). Localization of DMT1 is in agreement with the known

physiological site for ferrous iron absorption in the intestine, which is mostly restricted to

the brush border of the proximal intestine (84). An immunohistochemical study of

human duodenum showed that DMT1 localizes to enterocytes, especially at the

microvillus brush border membrane (85). Also, preincubation with DMT1 antibodies

significantly inhibited iron uptake in Caco2 cells at pH 5.5 (19). Mice with specific

inactivation of DMT1 gene in the intestine were born alive, but they rapidly developed

iron-deficiency anemia (86). The phenotype of the intestine-specific DMT1 knockout

mouse firmly establishes that DMT1 is the major iron transporter in the small intestine.

Function of DMT1 in Endosomal Iron Release

In Belgrade (b) rats, iron uptake from TF by erythropoietic cells is diminished and

globin synthesis is defective (76,87). A small decrease in endocytosis of TF, associated

with diminished iron uptake and increased iron release by exocytosis from b rats,

indicated that the defect lies in the metabolism of TF-iron after its endocytosis (88). It

has also been shown that diferric-TF is taken up into b reticulocytes, but iron is poorly

retained, and much is recycled to the extracellular space along with TF, meaning that b

reticulocytes are unable to move iron out of the vesicle after endocytosis (89).

Furthermore, an apparent deficit was also observed with NTBI uptake into b rat

erythroid cells, indicating that the b defect is not simply due to a failure to dissociate iron

from TF (90). Given the evidence that endosomal iron transport is not efficient in the b

rat and both b rats and mk mice have the identical G185R mutation in DMT1, the

hypothesis was formed that DMT1 is the TF cycle endosomal iron transporter in addition

to an intestinal iron transporter (74).







Iron deficient
Fe-S Cluster
X
Active----
5' UTR IRE
Translation control


Iron sufficient


Non-active


3'UTR IRE
mRNA stability control


RMm

t


- -)< '
L t


- wia..4



- l


Figure 1-3. Overview of the iron-responsive element/iron-regulatory protein (IRE/IRP)
network. IRPs interact with IREs to coordinate the expression of proteins
involved in iron metabolism. IRP binding to IREs located at 5' untranslated
regions (UTRs) inhibits translation, whereas IRP binding to 3' UTR IREs
increases mRNA stability. Cellular iron loading turns IRP1 from its IRE-
binding form to the Fe-S cluster-containing inactive form. Low iron levels
promote accumulation of active IRP1, resulting in its binding to IREs.


i


.. f


1\









in SFM to deplete cells of TF. For uptake, cells were incubated with 2 pM 59Fe-ferric

citrate for 60 min, followed by three washes of cell-impermeant iron chelator solution to

remove surface-bound iron. Cell-associated radioactivity was determined by using a y-

counter. Uptake was expressed as cpm/mg protein.

Genetic Knock-in to Tag Endogenous ZIP14 of HepG2 Cells with 3xFlag Epitope

AAV knock-in vector construction. The epitope tagging strategy and method

have been described in detail previously (145,146). Briefly, about 1 kb of left and right

homologous arms of ZIP14 were PCR amplified by using Platinum Taq DNA

polymerase High Fidelity (Invitrogen) and genomic DNA isolated from HepG2 cells. The

primers for the knock-in (Table 2-1) were designed specifically for sequences upstream

or downstream of the stop codon. Both arms were inserted into the rAAV-Neo-Lox P-

3xFlag knock-in vector (Kindly provided by Dr. Zhenghe Wang, Case Western Reserve

University) by using the USERTM enzyme system (NEB). The entire USER-treated

reaction mixture was used to transform chemically competent E. coli (Escherichia coli)

by heat shock at 420C.

AAV-Cre expression vector construction. To make the AAV-Cre-encoding

vector, the Cre gene coding sequence was PCR amplified from pBS185 plasmid

(Addgene) by using gene-specific primers linked with Not I restriction sites at each end

(Table 2-1), and then ligated into pAAV vector (Stratagene).

Packaging of AAV viruses. Targeting viruses were packaged into HEK 293-AAV

cells by using AAV Helper-Free System (Stratagene) according to the manufacturer's

instructions. Viruses were harvested after 72h.

Infection of HepG2 cells with AAV virus and Screening for targeting clones.

HepG2 cells growing in T25 flasks were infected with targeting virus and G418-









internalization signal. To examine the role of this YSDI motif, mutants of tyrosine or

other amino acids in this motif need to be made.


103









LIST OF ABBREVIATIONS

AAV Adeno-associated virus

BSA Bovine serum albumin

bp Base pair

Cp Ceruloplasmin

Dapi 4',6-diamidino-2-phenylindole

Dcytb Duodenal cytochrome B

DMEM Dulbecco's modified Eagle's medium

DMT1 Divalent metal transporter 1

EPO Erythropoietin

ER Endoplasmic reticulum

FPN1 Ferroportin 1

GFP Green fluorescence protein

GI Gastrointestinal

Hb Hemoglobin

HCP1 Heme carrier protein 1

HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

HFE Gene mutated in the most common form of hemochromatosis

HH Hereditary hemochromatosis

Hp Haptoglobin

HO Heme oxygenase

HRP Horseradish peroxidase

Hx Hemopexin

Irt Iron-regulated transporter

IRP Iron-regulatory protein









Total cell lysate
- + -
- +


. mr- n


Cell-surface protein


+ -
- +


50 KDa


anti-TFR2P


anti-Nat, K+ ATPase


i- -- n-
O no^"iliB .ls": 'q^^I I


SIlll 100 KDa

S .. 100 KDa.. .
-100 KDa


anti-Tubulin


S50 KDa


Figure 3-8. Effect of holo-TF on the abundance and cell-surface expression of ZIP14 in
HepG2 cells. HepG2 cells were incubated overnight in serum-free medium
supplemented with 10 pM apo-TF or 10 pM holo-TF. Total cell lysates and
cell-surface proteins were obtained and analyzed by Western blotting for
ZIP14, TFR2, Nat, KATPase, and tubulin. TFR2 was used as positive
control for treatment with holo-TF. Nat, K' ATPase and tubulin were used as
markers for plasma membrane and cytosolic proteins, respectively. The
results shown are representative of one of four experiments without
significant variation between experiments.


Holo-TF:
Apo-TF:
anti-Flag
(ZIP14)





I









pathway saturates at low TF concentrations between 50-100 nM (108,109). Here it has

been documented that in HepG2 cells incubated with 100 nM TF, knockdown of

endogenous ZIP14 resulted in 45% less uptake of iron from TF compared to controls.

Herbison et al. (176) showed that knockdown of TFR1 in HuH7 cells resulted in

80% less uptake of iron from 50 nM TF compared to controls. The reduced uptake was

due to diminished uptake of TF, because TF uptake was also 80% lower. In our study,

knockdown of ZIP14 in HepG2 cells did not affect the uptake of TF, but it did result in a

45% lower uptake of iron. These data provide strong support that ZIP14 plays a role in

the uptake of iron from TF, presumably through the high-affinity, low-capacity TFR1-

mediated endocytic pathway.

It was further shown that ZIP14 in HepG2 cells localizes not only on the plasma

membrane, but also to the endosome, where it partially colocalized with internalized TF.

Competition between TBI and NTBI uptake by hepatocytes indicates that a common

iron carrier participates in both pathways. A prerequisite for this is that the protein

involved should have a plasma membrane and intracellular distribution. Subcellular

localization study has shown that DMT1 in human hepatocytes located mostly in the

cytoplasm but faintly in the cell membrane (138). In contrast, in this study ZIP14 has

been shown to distribute widely on the plasma membrane and intracellularly. Moreover,

DMT1 is hardly detectable in hepatocytes. Taken together, these results strongly

indicate that ZIP14 may play an important role in TBI uptake in hepatocytes. I also

found that holo-TF had no effect on the abundance of ZIP14 in HepG2 cells. From this

observation, it is predicted that hepatic ZIP14 levels would not be altered in response to

elevated plasma concentrations of holo-TF, such as in hereditary hemochromatosis.










Basolateral


Fe3s Fes+
Diferric-TF


Figure 1-2. Schematic model for intestinal iron absorption. Dietary non-heme iron
(mostly Fe3+) is reduced by a ferric reductase (likely Dcytb) to yield ferrous
iron (Fe2+), which subsequently enters the enterocytes via DMT1. Heme is
absorbed intact, perhaps via heme carrier protein 1 (HCP1), and is then
catabolized by heme oxygenase 1 (HO) to liberate Fe2 If body iron stores
are high, iron may be stored in ferritin and lost through sloughed epithelial
cells within 3-4 days. Iron efflux across the basolateral membrane into
circulation is through FPN1. The exiting iron is oxidized to Fe3+ by hephaestin
to enable loading onto TF.


Apical









reaction buffer, and Pfu DNA polymerase. The mixture was amplified by polymerase

chain reaction. Initially the reaction mix was incubated at 95 oC for 30 s. Cycles were as

follows: denaturation for 30 s at 95 oC, annealing for 1 min at 55 oC, and extension

synthesis at 68 oC for 7 min for 18cycles. PCR products were digested with Dpnl

enzyme to remove the parental strands. The digested DNA mixture was transformed

into E. coli XL1-blue cells by heat shock at 420C. Mutagenesis products were all verified

by DNA sequencing.

Measurement of Iron Transport Activity

NTBI uptake was measured as previously described (129). Briefly, transfected

HEK 293T cells were washed three times in SFM and then incubated for 2 h in SFM to

deplete cells of TF. For uptake, cells were incubated with 2 pM 59Fe-ferric citrate for 2 h,

followed by three washes of cell-impermeant iron chelator solution to remove surface-

bound iron. Cell-associated radioactivity was determined by using a y-counter. Uptake

was expressed as cpm/mg protein.

Statistical Analysis

Data represent mean SEM. The TBI uptake studies were analyzed by unpaired

Student's t-test (GraphPad Prism, GraphPad Software). Iron transport activity studies of

the asparagine mutants were analyzed by one-way analysis of variance and Tukey's

post-hoc test (GraphPad Prism, GraphPad Software). The pH-dependence of iron

uptake data were analyzed as a completely randomized block design using the Glimmix

procedures (SAS Inst. Inc.). Blocks were based on the replication of the experiment.

Fixed effect included treatment and pH as well as the interactive effects of treatment

and pH. Block was the random effect. Multiple comparison were adjusted by Tukey-











Extracellular


Intracellular


Figure 1-6. Schematic demonstration of mutations in the iron transport protein DMT1.
Identified DMT1 mutations in rodents (A G185R) and human (* V114 del,
G212V, E399D and R416C).









4-6 Schematic illustration of mZipl4 membrane topology. ................................... 95

4-7 N-linked glycosylation does not affect plasma membrane localization of
m Z ip l4 ................................................................................... 9 6

4-8 Functional analysis of mZipl4 lacking N-linked glycosylation ...........................97









the total cells in the liver (107). They can acquire iron through both TBI and NTBI

pathways (Fig. 1-7).

Transferrin-bound Iron Uptake by Hepatocytes

In normal situations, the majority of circulating iron is bound to TF. The uptake of

TF-bound iron by hepatocytes from plasma is mediated by transferring receptor. In

addition, a receptor-independent TBI uptake pathway has also been proposed to exist in

hepatocytes (108,109). By using perfused liver, isolated primary hepatocytes or

hepatoma cell lines, it has been shown that there is a saturable high-affinity TBI uptake

described as a TFR1-dependent pathway and a nonsaturable low affinity process for

TBI uptake which is independent of TFR1 (108,110,111). The uptake of holo-TF through

TFR1 is a highly regulated process and most regulation occurs at the level of

posttranscriptional mRNA stability. Iron responsive elements (IREs) are present in the 3'

UTR of TFR1 mRNA (Fig. 1-3). When intracellular iron content is low, IRPs bind to the

IRE, protecting the mRNA from endonuclease cleavage. Consequently more TFR1

protein is synthesized. Accordingly, in iron-deficient animals or humans, TFR1 levels

are higher in the liver and other tissues. During iron excess, IRPs are inactivated by

iron-sulfur cluster formation, thus inhibiting binding to IREs, resulting in increased

degradation of TFR1 message and protecting cells from accumulating more iron (112).

The high-affinity process becomes saturated at relatively low extracellular TF

concentrations (50 to 100 nM) (108,109). It is likely that the receptor-independent low

affinity pathway predominates at normal physiological level of plasma TF (25 pM to 50

pM) (113).

Under certain pathological circumstances when the iron carrying capacity of TF

becomes exceeded, such as hemochromatosis, TFR1 is down-regulated in










+ PNGase F


-50 KDa







PNGase F





50 KDa


PNGase F




- 50 KDa


Figure 4-4. Deglycosylation of Zipl4. A) HEK 293T cells were transiently transfected
with PCMVSport6/mZipl4. B) HEK 293T cells were transiently transfected
with N-T-Flag-mZipl4. C) HepG2-ZIP14-Flag cell lysates. Five micrograms of
protein were incubated with or without PNGase F before Western analysis.
Overexpression of mZipl4 in HEK 293T cells results in two bands by Western
blot. The upper band disappears with PNGase F treatment, indicating the
higher molecular mass band is N-linked glycosylated form of mZipl4, while
the lower band represents the deglycosylated form. Endogenous ZIP14 is
also glycosylated as the ZIP14 band detected by Flag antibody shifted down
with PNGase F treatment.


anti-Zipl 4


anti-Flag
Zipl4


anti-Flag
ZIP14









Diferric TF

Y









Endosome 0..,/
SDMT1 DMT






5e o > Mitochondria


Ferritin

Cytoplasm



Extracellular

Figure 1-4. Overview of the TF cycle. When diferric TF binds TFR1 on the cell surface,
the complex internalizes through receptor-mediated endocytosis. Endosomes
become acidified by a proton pump. Acidification leads to protein
conformational changes that cause iron to dissociate from TF. Fe3+ released
from TF is reduced by ferrireductase STEAP3, and is then transported out of
the endosome, presumably through DMT1. TF and TFR1 both return to the
cell surface, where they separate at neutral pH. Both proteins participate in
further rounds of iron delivery.









CHAPTER 2
MATERIALS AND METHODS

Cell Culture

All cells were maintained in an incubator at 37C and 5% CO2. HEK 293T cells

were grown in Dulbecco's Modified of Eagle's Medium (DMEM, Mediatech) with 4.5 g/L

glucose, 4 mM L-glutamine, 1 mM sodium pyruvate, 100 U/ml penicillin, 100 [tg/ml

streptomycin and 10% fetal bovine serum (FBS, Atlanta Biologicals). HepG2 cells were

maintained in DMEM with 4.5 g/L glucose, 4 mM L-glutamine, 1 mM sodium pyruvate,

1x MEM nonessential amino acids (Mediatech), 100 U/ml penicillin, 100 [tg/ml

streptomycin and 10% FBS.

Expression of Mouse Zipl4 (mZipl4) in HEK 293T and HepG2 cells

Effectene reagent (Qiagen) was used for transient transfection of HEK 293T cells.

Briefly, HEK 293T cells were seeded at 40% confluency in 6-well plates. Transfection

began 24 h after seeding with 0.4 pg of plasmid DNA, 3.2 pl of enhancer and 10 pl of

Effectene reagent, and was carried out for 48 h before further analysis. For HepG2 cell

transfection, JetPei-Hepatocyte transfection reagent (Genesee Scientific) was used.

HepG2 cells were seeded at 40% confluency in 6-well plates. Twenty-four hours later, 3

pg of plasmid DNA, 9.6 pl of JetPei-Hepatocyte were diluted in 100 pl of 150mM NaCI

solution and mixed together. The mixture was incubated at room temperature for 15 min

before added to each well.

Knockdown of Endogenous ZIP14 in HepG2 Cells Using siRNA

SMARTpool siRNA (small interfering RNA) specific for human ZIP14 (Genbank

accession no. NM_015359) and non-targeting pool negative control siRNA were

purchased from Dharmacon (Thermo Scientific). Lipofectamine RNAiMAX transfection









Results

Overexpression of Zipl4 Increases TBI Uptake in HEK 293T Cells

HEK 293T cells, an easily transfectable cell line, were used to investigate the

effect of Zipl4 overexpression on TBI uptake. I found that HEK 293T cells transfected

with mZipl4 took up 25% more 59Fe-TF than did control cells transfected with empty

vector (Fig. 3-1A). Western blot analysis confirmed that the enhanced uptake of 59Fe-TF

was associated with higher Zipl4 protein levels (Fig. 3-1B). I also measured levels of

TFR1 and DMT1, which may also function in this iron uptake pathway. Levels of these

proteins did not differ between cells overexpressing Zipl4 and controls (Fig. 3-1 B).

These results suggest that Zipl4 may play a role in TBI uptake.

pH-dependent Iron Transport Activity of Zipl4

To determine the pH dependence of Zipl4-mediated iron transport, I transfected

HEK 293T cells with Zipl4 or empty vector, and measured the uptake of 59Fe-ferric

citrate by cells incubated in medium at pH 7.5, 6.5, or 5.5. I found that cells

overexpressing Zipl4 took up more iron than did controls at pH 7.5 and 6.5, but not at

5.5 (Fig. 3-2).

Subcellular Localization of Zipl4-GFP in HepG2 Cells

I used confocal laser-scanning microscopy to examine the subcellular distribution

of Zipl4-GFP in HepG2 cells after transient expression. Zipl4-GFP was readily

detectable at the hepatocyte plasma membrane and displayed abundant punctuate

intracellular staining that partially colocalized with internalized Texas-red labeled TF, a

marker of recycling endosomes (Fig. 3-3). Localization of Zipl4-GFP to the plasma

membrane is consistent with its postulated role in the uptake of NTBI at the cell surface.












































Figure 3-6. Confocal microscopic analysis of the subcellular localization of ZIP14 in
HepG2 cells. A) Staining of nuclei using DAPI. B) ZIP14 is detected at the
plasma membrane and in intracellular puncta. Endogenous ZIP14 in HepG2-
ZIP14-3xFlag cells was detected in permeabilized cells by using anti-Flag
antibody followed by rhodamine-conjugated secondary antibody. C) Detection
of internalized holo-TF. Cells were incubated for 30 min with Alexafluor 488-
labeled human holo-TF prior to fixation and permeabilization. D) Merged
image of panels A-C to visualize colocalization of ZIP14 and TF. E) Areas of
colocalization (designated by white) as determined using the colocalization
tool provided with the Leica-SP5 software. All images were obtained by using
a Leica TCS SP5 laser-scanning confocal microscope.




77









Other heme proteins
Hemoglobin
Haptoglobin Hemopexin


Heme ,.

0 w


U$


CD163 LRPCD91


SEndocytosis -
Receptor recycling \ Receptor recycling

Lysosome / .,

Lyo m Degraded ligand protein
Heme

Bilirubin CO Fe


Figure 1-5. Overview of the receptor-mediated pathway for endocytosis of extracellular
heme and hemoglobin. CD163 and LRP/CD91 represent two pathways for
uptake of extracellular heme incorporated in haptoglobin-hemoglobin (Hb-Hp)
and heme-hemopexin (Heme-Hx) complexes. Both receptors are highly
expressed in phagocytic macrophages, which can metabolize heme into
bilirubin, Fe and carbon monoxide. In addition to the expression in
macrophages, LRP/CD91 is also highly expressed in several other cell types
including hepatocytes and neurons.









33. Wallander, M. L., Leibold, E. A., and Eisenstein, R. S. (2006) Biochim Biophys
Acta 1763, 668-689

34. Park, C. H., Valore, E. V., Waring, A. J., and Ganz, T. (2001) J. Biol. Chem. 276,
7806-7810

35. Nemeth, E., and Ganz, T. (2006) Annu. Rev. Nutr. 26, 323-342

36. Cohen, L. A., Gutierrez, L., Weiss, A., Leichtmann-Bardoogo, Y., Zhang, D. L.,
Crooks, D., Sougrat, R., Morgenstern, A., Galy, B., Hentze, M. W., Lazaro, F. J.,
Rouault, T. A., and Meyron-Holtz, E. (2010) Blood

37. Torti, F. M., and Torti, S. V. (2002) Blood 99, 3505-3516

38. Munro, H. N., Aziz, N., Leibold, E. A., Murray, M., Rogers, J., Vass, J. K., and
White, K. (1988) Ann. N. Y. Acad. Sci. 526, 113-123

39. Harrison, P. M., and Arosio, P. (1996) Biochim. Biophys. Acta 1275, 161-203

40. Muckenthaler, M. U., Galy, B., and Hentze, M. W. (2008) Annu. Rev. Nutr. 28,
197-213

41. Aisen, P., and Listowsky, I. (1980) Annu. Rev. Biochem. 49, 357-393

42. Nunez, M. T., Gaete, V., Watkins, J. A., and Glass, J. (1990) J. Biol. Chem. 265,
6688-6692

43. Kuhn, L. C., McClelland, A., and Ruddle, F. H. (1984) Cell 37, 95-103

44. Jing, S. Q., and Trowbridge, I. S. (1987) EMBO J. 6, 327-331

45. Andrews, N. C. (2000) Annu. Rev. Genomics Hum. Genet. 1, 75-98

46. Yamashiro, D. J., Tycko, B., Fluss, S. R., and Maxfield, F. R. (1984) Cell 37, 789-
800

47. Sipe, D. M., and Murphy, R. F. (1991) J. Biol. Chem. 266, 8002-8007

48. Sendamarai, A. K., Ohgami, R. S., Fleming, M. D., and Lawrence, C. M. (2008)
Proc Natl Acad Sci U SA 105, 7410-7415

49. Chan, L. N., and Gerhardt, E. M. (1992) J. Biol. Chem. 267, 8254-8259

50. Bernstein, S. E. (1987) J. Lab Clin. Med. 110, 690-705


106









LIST OF FIGURES


Figure page

1-1 Distribution of iron w within the body ......... ....................................... .................. 44

1-2 Schematic model for intestinal iron absorption. ............................... ................ 45

1-3 Overview of the iron-responsive element/iron-regulatory protein (IRE/IRP)
network. ......................................... ........ ........................... 46

1-4 O verview of the TF cycle. ................. ......... ................................................47

1-5 Overview of the receptor-mediated pathway for endocytosis of extracellular
h e m e a nd h e m o g lo b in ................ ................... ............................................ 4 8

1-6 Schematic demonstration of mutations in the iron transport protein DMT1.........49

1-7 Iron uptake by hepatocytes ......... ............................ ....................... 50

3-1 Overexpression of Zip14 increases TBI uptake. ............. ......... ................ 73

3-2 pH dependence of Zipl4-mediated iron transport. ..........................................74

3-3 Subcellular localization of Zip14-GFP in HepG2 cells. .................................75

3-4 Comparison of ZIP14 and DMT1 mRNA levels in HepG2 and HEK 293T
ce lls ... ................................................................................. 75

3-5 Targeted knock-in of 3x Flag into the ZIP14 locus of HepG2 cells.......................76

3-6 Confocal microscopic analysis of the subcellular localization of ZIP14 in
HepG2 cells. .............. .. ....... ...... ..... ................... 77

3-7 Knockdown of ZIP14 decreases TBI uptake by HepG2 cells...................................78

3-8 Effect of holo-TF on the abundance and cell-surface expression of ZIP14 in
HepG2 cells. ............. ......... ....................... ...........79

4-1 Schematic representation of the mZipl4 gene and sequence alignment of its
tw o protein isoform s. ........................ .................................. ............. 90

4-2 Flag epitope m apping of mZip14. .......... .. .......................................... ...... .. 91

4-3 The carboxy terminus of mZip14 is intracellular ............................................ 92

4-4 D eglycosylation of Z ipl4........... .......................................... ......... ...... 93

4-5 Identification of N-linked glycosylation sites in mZip14 ................... ................ 94










Total cell lysate


anti-Flag
(Zip14)



anti-LAMP1


anti-Na+,K+
ATPase


WT mZipl 4


Cell-surface protein

tp -^


50 KDa


100 KDa


100 KDa


ASNs mutant


Figure 4-7. N-linked glycosylation does not affect plasma membrane localization of
mZipl4. A) HEK 293T cells were transiently transfected with empty
pCMVSport2 vector (Sport2), wide-type N-T-Flag-mZip4 (WT mZipl4) or its
mutant with all four glycosylated asparagines replaced with aspartic acids
(ASNs mutant). After 24 h, total cell lysates were harvested by using SDS
lysis buffer, while cell-surface proteins were labeled with NHS-SS-biotin.
Samples were analysis by Western blotting for Zipl4. After stripping, blots
were reprobed for Lamp1 and Na, K ATPase as markers for cytosolic and
plasma membrane proteins, respectively. B) After transfection, both WT and
ASNs mutant mZipl4 are detected at the plasma membrane in
nonpermeabilized cells. Zipl4 signals were detected by using anti-Flag
antibody followed by rhodamine-conjugated secondary antibody (red). Nuclei
were stained with Dapi (blue). Data represent three independent experiments.









Murine models reveal the important role of Tf/TfR1-mediated iron uptake for

erythropoiesis. Firstly, hypotransferrinemic (Trfpx/hpx) mice carry a spontaneous

mutation in the Tf gene (50). These mice are Tf deficient, with only about 1% of normal

circulating Tf concentrations. They develop severe microcytic hypochromic anemia,

indicating an indispensable role of Tf in iron delivery to developing erythroid cells (51).

Knockout mice have also been produced which lack functional TfR1 protein (52). These

mice die in utero between embryonic day 9.5 and 11.5, apparently as a consequence of

severe anemia. The heterozygous mice lacking one copy of the TfR1 gene demonstrate

iron-deficient erythropoiesis, even though they retain one normal TfR1 allele.

Iron Recycling

The major source of plasma iron does not come from intestinal absorption, but

from macrophages that recycle iron from senescent or damaged erythrocytes (53).

Macrophage iron recycling is quantitatively important because the amount of iron

supplied by this way each day is about 20 times greater than the amount absorbed

through the small intestine (2). Iron recycling takes place in the RES, a term that

describes specialized macrophages that are found in the liver (Kupffer cells), the spleen

and the bone marrow (53). Binding of erythrocytes to the macrophage cell surface

initiates phagocytosis and lysosome-mediated degradation of effete RBCs. Heme is

liberated and catabolized by HO, releasing inorganic iron, biliverdin and carbon

monoxide (54). Free iron is either stored in ferritin or released into the circulation

through the iron export protein FPN1. FPN1 is highly expressed in extraembryonic

visceral endoderm cells which are important for providing nourishment to the developing

embryo. Global inactivation of the murine FPN1 gene leads to embryonic lethality due to

a defect of the developing embryo to acquire iron, indicating that FPN1 is essential in









ACKNOWLEDGMENTS

Four years of PhD study is the time I learned most and the writing of this

dissertation has been one of the most significant academic challenges I have ever had

to face. Without the support and help of the following people, this project would not have

been completed. It is to them that I own my deepest gratitude.

Dr. Mitchell D. Knutson, who is my major professor. He led me into molecular

nutrition field and taught me how to be a better scientist. He has always been a

generous, tireless and constant mentor and supporter. His wisdom, knowledge and

commitment to the higher standards inspired and motivated me.

Dr. Harry S. Sitren, Dr. Bobbi Langkamp-Henken and Dr. Lei Zhou, who are my

dissertation committee members. They have provided invaluable support to me over the

years. Their advice and guidance for my dissertation work were highly valued and

appreciated.

Hyeyoung Nam, Supak Jenkitkasemwong, Chia-Yu Wang, Charlie Michaudet and

Stephanie Duarte, who are my colleagues and friends. They always help me with

various aspects of my experiments and provide valuable advice for my work.

I would also like to thank my parents who always support, encourage and believe

in me. I sincerely thank my wife, Miaomiao Wu, for your love, support, encouragement

and patience.









N-linked Glycosylation does not Affect Plasma Membrane Localization of mZipl4

To determine the role of N-linked glycosylation in its trafficking to the plasma

membrane, HEK 293T cells were transiently transfected with wild-type (WT) N-T-Flag

mZipl4 (WT mZipl4) and its ASNs mutant (which has all four glycosylated asparagines

mutated to aspartic acids). Total cell lysates and cell-surface proteins biotinylated with a

membrane impermeable reagent NHS-SS-biotin were analyzed by Western blotting. In

total cell lysate, WT mZipl4 exhibited two bands whereas the ASNs mutant resulted in

only one band of the lower molecular mass. A strong signal of the mutant could also be

detected at the cell surface (Fig.4-7A). The subcellular localization of the WT mZipl4

and ASNs mutant were examined by confocal microscopy using the antibody against

the Flag epitope under nonpermeabilized conditions. Both wide-type and ASNs mutant

could be detected on the plasma membrane (Fig. 4-7B). Cells transfected with empty

vector had no detectable immunofluorescence (Data not shown). It was concluded that

N-linked glycosylation of mZipl4 is not required for the efficient transport of the protein

to the cell surface, at least in the HEK 293 overexpression system.

N-linked Glycosylation is Required for the Iron Transport Activity of mZipl4

It was found that the iron transport activity decreased significantly in the ASNs

mutant-transfected cells compared to that of WT mZipl4-transfected cells (Fig 4-8A).

Western analysis indicated that the protein expression level of the mutant was similar to

WT mZipl4 (Fig 4-8B).

Discussion

ZIP proteins have been predicted to have eight-TM domains (130,178). However,

this does not apply to all the ZIPs. Human ZIP14 and ZIP8 have been predicted to be

different from other ZIPs in that only one transmembrane (TM) domain exists instead of









iron metabolism is also supported by the evidence that hepatocytes from Hfe'- mice can

take up more NTBI compared to wide-type mice (141). However, the mechanism by

which HFE regulates iron metabolism still remains to be determined. Interestingly,

overexpression of HFE in HepG2 cells decreased ZIP14 levels by decreasing the

stability of ZIP14 (136). The reduced ZIP14 levels were associated with diminished

uptake of not only NTBI but also TBI, suggesting that ZIP14 participates in both

pathways of iron acquisition.

A study in isolated rat hepatocytes has shown that LPS markedly increases the

uptake of TBI (142). Similarly in HepG2 cells, stimulation with the inflammatory cytokine,

IL-6, enhanced the uptake of TBI by 48%(143). LPS and IL-6 have both been shown to

potentially increase levels of Zip14 in mouse liver and in isolated hepatocytes (134).

These observations suggest that ZIP14 is involved in TBI uptake by hepatocytes.

In summary, ZIP14 is a newly identified iron import protein. It localizes to the

plasma membrane and mediates the uptake of NTBI into cells. Expression of Zipl4 was

induced by both LPS and IL-6, which could also stimulate the TBI uptake in

hepatocytes. HFE expression decreased ZIP14 stability and reduced the uptake of both

NTBI and TBI in HepG2 cells. The above evidence suggests that ZIP14 is a common

transporter shared by both NTBI and TBI. The research described herein tested the

hypothesis that ZIP14 plays a role in the uptake of TBI. Studies were also directed to

investigate the membrane topology of ZIP14 as a first step to understand which

structural elements contribute to the iron transport ability of ZIP14.









hepatocytes. In untreated hereditary hemochromatosis (HH) patients the expression of

TFR1 was undetectable in hepatocytes (114).

TFR1 is not the only transferring receptor in hepatocytes. TFR2, a homolog of

TFR1, is highly expressed in hepatocytes and in developing erythroid precursor cells,

and it may play a role in liver iron loading (115). Similar to the ubiquitously expressed

TFR1, TFR2 is a type-2 membrane protein with a cytoplasmic N-terminus, a single

transmembrane domain and a large ectodomain. The amino acid sequences of

transmembrane region and the extracellular domain in TFR2 shares 45% identity to that

of TFR1. TFR2 protein is up-regulated in iron overload and in a mouse model of HH and

may contribute to increased TBI uptake by the liver during iron overload conditions

(116). Mutations in TFR2 cause hemochromatosis. TFR2 can bind and internalize holo-

TF, but its affinity for holo-TF is 25-30 times lower than that of TFR1 (117,118). Unlike

TFR1, TFR2 lacks an IRE, the protein levels increase in response to increased level of

holo-TF (119,120). A study of TFR2-null mice indicated that TFR2 has a minor role in

iron transport and hepatic iron-loading (121).

Non-transferrin-bound Iron Uptake by Hepatocytes

The amount of circulating TF-bound iron is determined by three coordinated

process: macrophage iron recycling, duodenal iron absorption and hepatic iron

storage/release (122). As TF becomes saturated in iron overload states, excess iron is

also found as NTBI. NTBI will exist in high amount and contributes considerably to

hepatic iron loading (123,124). The form of NTBI present in the plasma could be bound

to either citrate or albumin (125). NTBI is likely to play an important role in hepatocyte

iron loading in HH and other iron overload conditions. NTBI is cleared rapidly by the

liver from plasma, demonstrated by cases of human and mouse congenital TF










Table 4-1. Bioinformatic Prediction of mZipl4 Transmembrane (TM) Regions

Programs Numberof Predicted TMs Predicted Predicted
TMs N C
1-150 151-300 301-450 450-
TMHMM 7 Outside 151-173 180-202 217-234 349-371 391-413 420-442 457-479 Inside
HMMTOP 7 Outside 156-175 184-202 217-234 349-371 396-415 424-442 467-484 Inside
TopPred 7 Inside 4-24 153-173 184-204 217-237 352-372 398-418 466-486 Outside
MINNOU 7 N/A 148-173 181-209 219-241 359-373 392-417 422-448 457-484 N/A
TmPred 8 Outside 4-22 151-169 184-202 217-238 346-372 395-415 420-439 462-482 Outside
8 Inside 4-24 151-173 184-202 216-233 346-372 395-415 420-439 462-482 Inside
ConPred II 8 Outside 4-24 153-173 185-205 216-236 351-371 396-416 422-442 462-482 Outside
MEMSAT-SVM 8 Outside 150-175 186-201 221-237 334-353 362-384 394-416 424-442 463-482 Outside
MEMSAT III 6 Outside 153-176 184-204 351-374 396-420 423-442 460-483 Outside
Phobius 6 Outside 151-173 185-205 217-234 396-415 421-438 458-482 Outside
SOSUI 4 Outside 4-24 151-173 186-208 216-238 Outside


Table 4-2. Prediction of mZipl4 Signal Peptide Cleavage Site


Programs

SignalP 3.0
SignalP V2.0.b2
SPEPlip
Phobius
Predisi
Sig-Pred

SOSUI
MEMSAT-SVM


Existence of
Signal Peptide
Y
Y
Y
Y
Y
Y
Y
Y
Y


Cleavage
Position
28/29
28/29
28/29
28/29
24/25
24/25
22/23
21/22
16/17


Cleavage
Site
A/S
A/S
A/S
A/S
P/Q
P/Q
T/A
R/T
L/F









CHAPTER 3
ZIP14 AND TRANSFERRIN-BOUND IRON UPTAKE

Introduction

Iron is essential for almost all known organisms. Iron uptake by cells is a carrier-

mediated process and is primarily through the TF-TFR1 complex, a process known as

TBI uptake. This receptor-mediated endocytosis process involves the transfer of iron out

of endosomes into the cytosol. Normally, iron is transported in the plasma bound to TF.

But when the iron carrying capacity of TF becomes exceeded during conditions of iron

overload, such as in hemochromatosis and thalassemia, NTBI may present in high

quantities. Thus, TBI and NTBI will coexist in the plasma (160,161). Under normal

conditions, the liver takes up TBI, almost exclusively into hepatocytes (162). The uptake

of NTBI into hepatocyte cell lines is mediated, at least in part, by the transmembrane

protein ZIP14, a member of the ZIP family of metal-ion transporters (129,136). At least

three studies have shown that TBI and NTBI compete for uptake by hepatocytes

(160,163,164), suggesting the existence of transporters) shared by these two

pathways. Up to now, DMT1, a major ferrous transporter for dietary absorption, is the

only known transport protein which can function in endosomal iron release.

However, at least three lines of evidence support the existence of other iron

transporters) involved in TBI uptake by cells. Firstly, DMT1 knockout (DMT1-'-) mice

were born alive (86), indicating fetal DMT1 is not required for maternofetal iron

transport, which relies primarily on TBI delivery (165,166). Secondly, in Belgrade rats,

which have a loss-of-function mutation in DMT1, TBI uptake was still effective in

duodenal crypt cells, which depend mainly on the TF cycle for iron uptake, indicating

that DMT1 is not required for the uptake of TBI by these cells (102). Thirdly, in the case









confirmed correct knock-in of the flag epitope at the C-terminus and insertion of the

NeoR gene. Excision of the NeoR gene and identification of a Flag allele were confirmed

by PCR using primers (P5 + P6) (Fig. 3-5B) and by DNA sequencing. Western blot

analysis using anti-Flag antibody revealed translation of a Flag-tagged protein (Fig. 3-

5C). The flag-immunoreactive band is detected between ~55 and 60 kDa, consistent

with the calculated molecular mass of ZIP14 (54 kDa) plus the 3xFlag (~ 3 kDa).

Moreover, the Flag-immunoreactive band could be knocked down by using siRNA

targeting ZIP14, thus confirming the band as ZIP14 (Fig. 3-5C).

Subcellular Localization of ZIP14 in HepG2 Cells

Previous immunofluorescence studies in primary mouse hepatocytes detected

Zip14 at the plasma membrane of non-permeabilized cells (134). Here I used

immunofluorescence analysis of HepG2 cells expressing endogenous ZIP14-3xFlag to

further investigate the subcellular localization of ZIP14 (Fig. 3-6). In permeabilized

HepG2 cells, I detected ZIP14-3xFlag in intracellular puncta, as well as on the plasma

membrane (Fig. 3-6B). The specificity of the anti-Flag antibody was confirmed by the

absence of immunofluorescent staining in permeabilized wild-type HepG2 cells (data

not shown). To determine if intracellular ZIP14 localizes to endosomes, I used

Alexafluor 488-labeled human holo-TF, which is endocytosed by the cells (Fig. 3-6C).

As shown in Fig. 3-6D, ZIP14 and the labeled TF partially colocalize in the cytosol. A

colocalization rate of 56% between ZIP14 and TF was calculated by using the

colocalization algorithm provided with the Leica-LSM SP5 software (Fig. 3-6E). The

presence of ZIP14 in TF-positive endosomes is consistent with the hypothesis that

ZIP14 plays a role in the uptake of iron from TF.









consensus sequence for the catalytic zinc-binding site of metalloproteases within ZIP

proteins), Taylor et al. (133) identified ZIP14, which they designated LZT-Hs4. Human

Zipl4 was predicted to be 53 kDa with up to three additional N-linked carbohydrate side

chains. Expression of human ZIP14 in CHO cells showed an apparent molecular mass

of 53 kDa. Under nonreducing conditions, ZIP14 migrated as a trimer. ZIP14 was

expressed on the plasma membrane in nonpermeabilized cells. Human ZIP14 was

predicted to have a long N terminus, followed by 8 putative transmembrane domains

and a short C terminus (129,132,133).

The metal transport activity of ZIP14 was first characterized in 2005 by Taylor et

al. (132), who showed that ZIP14 overexpression in CHO cells stimulated the uptake of

zinc into the cytosol. In 2005, Liuzzi et al. (134) found that transfection of mouse Zipl4

cDNA into HEK 293 cells increased zinc uptake and that Zipl4 was the most

upregulated zinc transporter in response to lipopolysaccharide (LPS) treatment or

turpentine-induced inflammation in the mouse liver. However, the hypozincemic

response was milder in interleukin-6 knockout (IL-6-'-) mice exposed to LPS than in wild-

type mice. IL-6'- mice displayed neither hypozincemia nor Zipl4 induction with

turpentine-induced inflammation. Immunohistochemical analysis showed that, in

hepatocytes, plasma membrane expression of Zipl4 increased in response to both LPS

and turpentine. IL-6 also increased expression of Zipl4 in primary hepatocyte cultures.

It was concluded that ZIP14 is a zinc importer upregulated by IL-6 and Zipl4 plays a

major role in the hypozincemia accompanying the acute-phase response to

inflammation and infection.









TABLE OF CONTENTS

page

ACKNOW LEDGMENTS.......... ........ ........ .......................................4

L IS T O F T A B LE S ...................... ................ ................................................. 8

L IS T O F F IG U R E S ...................... ................ ............. .. ................................ 9

LIST OF ABBREVIATIONS ............... ............................. ......... .. ....... 11

A B S T R A C T ............. ......... .. ............. .. ....................................................... 1 4

CHAPTER

1 LITERATURE REVIEW ............... ................ ................ 16

Function and Distribution of Iron in the Body................................. 16
Iron H om eostasis .............................................................................. ... .... 17
Intestinal Iron Absorption ............................. .................... .....17
Intracellular Storage and Circulatory Transfer of Iron ........... ... ..............21
Iron Utilization by Erythrocytes ................................................. ........... 23
Iro n R e c y c lin g ................................................................................................. 2 4
Iro n S to ra g e ................................................... ..............................2 7
Iron Loss a n d E xcretion ........................................................... ........ .... ........ 2 7
R ole of D M T1 in Iron M etabolism ................................................................... 27
DMT1 Gene Structure and its Isoforms .................................. .......28
Role of DMT1 in Iron Absorption ....................................................... .... ............ 29
Function of DMT1 in Endosomal Iron Release ..................................... 31
Subcellular Localization of D M T 1 ...................................................................... 34
DMT1 is not the Only Transporter Involved in Endosomal Iron Release ........ 36
Iron Uptake by Hepatocytes .................................................................. 37
Transferrin-bound Iron Uptake by Hepatocytes.............................................. 38
Non-transferrin-bound Iron Uptake by Hepatocytes ....... .......................... 39
ZIP14 and Iron M etabolism .......................... ............. ....................................40
Identification and Characterization of ZIP14....................................... 40
ZIP14 and C cellular Iron U take .............. .. ............................................... 42

2 M AT ER IA LS A N D M ET H O D S ............. ............. ................................. ................51

C e ll C u ltu re ............. ......... ........... ... ... ................................ ...................... 5 1
Expression of Mouse Zipl4 (mZipl4) in HEK 293T and HepG2 cells.................51
Knockdown of Endogenous ZIP14 in HepG2 Cells Using siRNA ................. ...51
Measurement of TF-bound Iron Uptake .......................................... 52
pH Dependence of Iron Transport Activity ......................................................... 52
Genetic Knock-in to Tag Endogenous ZIP14 of HepG2 Cells with 3xFlag
E p ito p e ............................................................................ 5 3









51. Trenor, C. C., 3rd, Campagna, D. R., Sellers, V. M., Andrews, N. C., and
Fleming, M. D. (2000) Blood 96, 1113-1118

52. Levy, J. E., Jin, O., Fujiwara, Y., Kuo, F., and Andrews, N. C. (1999) Nat. Genet.
21, 396-399

53. Knutson, M., and Wessling-Resnick, M. (2003) Crit. Rev. Biochem. Mol. Biol. 38,
61-88

54. Ryter, S. W., and Tyrrell, R. M. (2000) Free Radic. Biol. Med. 28, 289-309

55. Donovan, A., Lima, C. A., Pinkus, J. L., Pinkus, G. S., Zon, L. I., Robine, S., and
Andrews, N. C. (2005) Cell Metab. 1, 191-200

56. Knutson, M. D., Vafa, M. R., Haile, D. J., and Wessling-Resnick, M. (2003) Blood
102, 4191-4197

57. Wilkins, M., Williams, P., and Cavill, I. (1977) Ann. Rheum. Dis. 36, 474-475

58. Tolosano, E., Fagoonee, S., Hirsch, E., Berger, F. G., Baumann, H., Silengo, L.,
and Altruda, F. (2002) Blood 100, 4201-4208

59. Nielsen, M. J., and Moestrup, S. K. (2009) Blood 114, 764-771

60. Kristiansen, M., Graversen, J. H., Jacobsen, C., Sonne, O., Hoffman, H. J., Law,
S. K., and Moestrup, S. K. (2001) Nature 409, 198-201

61. Lim, S. K., Kim, H., bin Ali, A., Lim, Y. K., Wang, Y., Chong, S. M., Costantini, F.,
and Baumman, H. (1998) Blood 92, 1870-1877

62. Hvidberg, V., Maniecki, M. B., Jacobsen, C., Hojrup, P., Moller, H. J., and
Moestrup, S. K. (2005) Blood 106, 2572-2579

63. Herz, J., and Strickland, D. K. (2001) J. Clin. Invest. 108, 779-784

64. Tolosano, E., Hirsch, E., Patrucco, E., Camaschella, C., Navone, R., Silengo, L.,
and Altruda, F. (1999) Blood 94, 3906-3914

65. Anderson, G. J., and Frazer, D. M. (2005) Semin. Liver Dis. 25, 420-432

66. Bacon, B. R., and Tavill, A. S. (1984) Semin. Liver Dis. 4, 181-192

67. Van Wyk, C. P., Linder-Horowitz, M., and Munro, H. N. (1971) J. Biol. Chem.
246, 1025-1031









Table 2-3. List of Primers Used for Generating Asparagine Mutants


Mutant Name Primer Sequence
N52D Forward 5' GGACCGCTATGGAAAGGATGACAGCCTTAC
CCT 3'
Reverse 5' AGGGTAAGGCTGTCATCCTTTCCATAGCGGT
CC 3'
N75D Forward 5' GAGTGGGCCGGGATGATGTTTCCCAGCC 3'
Reverse 5' GGCTGGGAAACATCATCCCGGCCCACTC 3'
N85D Forward 5' GGAAGGACCCAGGGACCTCTCCACGTG 3'
Reverse 5' CACGTGGAGAGGTCCCTGGGTCCTTCC 3'
N100D Forward 5' CTTTGCGGCGCACGACTTGAGCGAGCG 3'
Reverse 5' CGCTCGCTCAAGTCGTGCGCCGCAAAG 3'
N455D Forward 5' CCAGGAGGATGAGAAGGACGACAGCTTTCTG
GT 3'
Reverse 5' ACCAGAAAGCTGTCGTCCTTCTCATCCTCCT
GG-3'



























Figure 3-3. Subcellular localization of Zipl4-GFP in HepG2 cells. A) Staining of nuclei
DAPI. B) ZIP14-GFP is detected at the plasma membrane and in intracellular
puncta. C) Detection of internalized holo-TF. Cells were incubated for 30 min
with Texas red-labeled human holo-TF prior to fixation. D) Merged image of
panels B-C to visualize colocalization of Zip-GFP14 and TF. The arrows
indicate clear colocalization of Zipl4-GFP and TF. All images were obtained
by using a Leica TCS SP5 laser-scanning confocal microscope. Cells were
not permeabilized.

16
M HepG2
u I C HEK293
c, 12-





4
DZZ











Figure 3-4. Comparison of ZIP14 and DMT1 mRNA levels in HepG2 and HEK 293T
cells. Total RNA was isolated from untreated HepG2 and HEK 293T cells.
Transcript copy numbers were determined by using qRT-PCR. Data
represent the mean SEM of three independent experiments of triplicate
samples.
ix "-- 4









samples.









133. Taylor, K. M., and Nicholson, R. I. (2003) Biochim. Biophys. Acta 1611, 16-30

134. Liuzzi, J. P., Lichten, L. A., Rivera, S., Blanchard, R. K., Aydemir, T. B., Knutson,
M. D., Ganz, T., and Cousins, R. J. (2005) Proc. Natl. Acad. Sci. U. S. A. 102,
6843-6848

135. Grootveld, M., Bell, J. D., Halliwell, B., Aruoma, O. I., Bomford, A., and Sadler, P.
J. (1989) J. Biol. Chem. 264, 4417-4422

136. Gao, J., Zhao, N., Knutson, M. D., and Enns, C. A. (2008) J. Biol. Chem. 283,
21462-21468

137. Garrick, M. D., Kuo, H. C., Vargas, F., Singleton, S., Zhao, L., Smith, J. J.,
Paradkar, P., Roth, J. A., and Garrick, L. M. (2006) Biochem. J. 398, 539-546

138. Shindo, M., Torimoto, Y., Saito, H., Motomura, W., Ikuta, K., Sato, K., Fujimoto,
Y., and Kohgo, Y. (2006) Hepatol. Res. 35, 152-162

139. Olynyk, J. K., Cullen, D. J., Aquilia, S., Rossi, E., Summerville, L., and Powell, L.
W. (1999) N. Engl. J. Med. 341, 718-724

140. Feder, J. N., Gnirke, A., Thomas, W., Tsuchihashi, Z., Ruddy, D. A., Basava, A.,
Dormishian, F., Domingo, R., Jr., Ellis, M. C., Fullan, A., Hinton, L. M., Jones, N.
L., Kimmel, B. E., Kronmal, G. S., Lauer, P., Lee, V. K., Loeb, D. B., Mapa, F. A.,
McClelland, E., Meyer, N. C., Mintier, G. A., Moeller, N., Moore, T., Morikang, E.,
Prass, C. E., Quintana, L., Starnes, S. M., Schatzman, R. C., Brunke, K. J.,
Drayna, D. T., Risch, N. J., Bacon, B. R., and Wolff, R. K. (1996) Nat. Genet. 13,
399-408

141. Chua, A. C., Olynyk, J. K., Leedman, P. J., and Trinder, D. (2004) Blood 104,
1519-1525

142. Potter, B. J., Blades, B., McHugh, T. A., Nunes, R. M., Beloqui, O., Slott, P. A.,
and Rand, J. H. (1989) Am. J. Physiol. 257, G524-531

143. Hirayama, M., Kohgo, Y., Kondo, H., Shintani, N., Fujikawa, K., Sasaki, K., Kato,
J., and Niitsu, Y. (1993) Hepatology 18, 874-880

144. Worthington, M. T., Browne, L., Battle, E. H., and Luo, R. Q. (2000) Am. J.
Physiol. Gastrointest. Liver Physiol. 279, G1265-1273

145. Zhang, X., Guo, C., Chen, Y., Shulha, H. P., Schnetz, M. P., LaFramboise, T.,
Bartels, C. F., Markowitz, S., Weng, Z., Scacheri, P. C., and Wang, Z. (2008)
Nat. Methods 5, 163-165

146. Wang, Z. (2009) Methods Mol. Biol. 567, 87-98


112









W western Blot Analysis ..................................... ...... ........... .......... .. 54
Subcellular Localization ....................................... ........ .. ........... .. ... .... ... 55
Q uantitative Determ nation of TF uptake ...................................................... ... 55
Isolation of Plasma Membrane Proteins by Cell-surface Biotinylation ............. 56
RNA Isolation and Quantitative PCR (qPCR) ............................................... 56
Construction of GFP Fusion and Flag-tagged mZipl4 Plasmids.......................57
Primary Sequence Analysis and Bioinformatic Predictions..................................... 58
Construction of Asparagine Mutant Plasmids............... .... ................ 58
Measurement of Iron Transport Activity .......................................... ........ 59
Statistical A analysis ........................... .. .................... ................ 59

3 ZIP14 AND TRANSFERRIN-BOUND IRON UPTAKE.......................... ........... 63

Introd auction ............... ........... ......................... ........................... 63
Results ...................... .................... ........ ......... 65
Overexpression of Zipl4 Increases TBI Uptake in HEK 293T Cells.................. 65
pH-dependent Iron Transport Activity of Zipl4 ................ ............ ............... 65
Subcellular Localization of Zipl4-GFP in HepG2 Cells........ .............................65
ZIP14 and DMT1 mRNA Copy Numbers in HepG2 and HEK 293T Cells.........66
Epitope tagging of Human Endogenous ZIP14 in HepG2 Cells......................... 66
Subcellular Localization of ZIP14 in HepG2 Cells .................. ...................... 67
Knockdown of ZIP14 in HepG2 Cells Decreases TBI Uptake............................68
Effect of Holo-TF on the Abundance and Cell-surface Expression of ZIP14
in HepG2 Cells ................ ........................68
D is c u s s io n ......... ....................................... ........................... 6 9

4 STR UCTU RA L A NA LY SIS O F Z IP 14................................................. ... ................... 80

In tro d u c tio n ............. .. ............... ................. ............................................. 8 0
R e s u lts ............. .................. .... .. ...... ............................... ................ 8 3
Structure of the murine Zip14 Gene and Sequence Alignment of its Two
Protein Isoform s ........... ....................... ......... .. ................ ............... 83
Epitope Mapping of mZipl4........................................ .................. 83
Enzymatic Analysis of Glycosylation Status of Zipl4 ................... ......... ....... 84
Identification of N-linked Glycosylation Sites in mZipl4................................ 85
Schematic Membrane Topology Model of mZipl4 .............................................85
N-linked Glycosylation does not Affect Plasma Membrane Localization of
m Z ip l4 ............... ............. ........................ .............. .......... .... .......... 86
N-linked Glycosylation is Required for the Iron Transport Activity of mZip14.... 86
D discussion ................ ....................................... ...........................86

5 CONCLUSIONS AND FUTURE DIRECTIONS.................................................... 99

C conclusions ............... ........... ........................ ........................... 99
Future D directions ............... ............. .. ......... .....................101









heme can be cleared through these two receptor-mediated endocytosis pathways, thus,

avoiding the strong oxidative features and proinflammatory effects of free heme.

Iron Storage

The liver is the main storage organ for iron and accounts for approximately 50% of

the storage iron (65,66). In rats, about 98% of iron stored in the liver is found in hepatic

parenchymal cells and most of the remainder is found in Kupffer cells, with only very

small amounts in stellate cells, endothelial cells and bile duct cells (67).

Iron is primarily stored in macrophages of the RES and in hepatocytes of the liver.

Iron released from metabolized heme can be stored in two forms within the cell: as

ferritin in the cytosol and as hemosiderin after ferritin breakdown in lysosomes (8).

Hemosiderin in macrophages increases dramatically in iron overload, but only

represents a small portion of normal body iron stores. Tissue macrophages, particularly

in the spleen and liver, also express ferritin, which can store iron that is not needed

elsewhere.

Iron Loss and Excretion

Although iron homeostasis is tightly controlled in terms of uptake, recycling and

utilization, iron loss and excretion are not actively regulated. Iron is lost from the body

by sloughing of mucosal cells, desquamation of skin cells, blood loss, sweat and urinary

excretion. The amount of iron lost every day is about 1 to 2 mg, equivalent to the

amount taken up through normal daily diet (6,68).

Role of DMT1 in Iron Metabolism

Divalent metal transporter 1 (DMT1 or SLC39A2-solute carrier family 11,

member 2) is the most well-characterized transmembrane iron transport protein.









Ferritin is a 450 kDa, cage-like heteropolymer of 24 subunits, including H- (heavy,

21 kDa) chains and L- (light, 19 kDa) chains (38). This structure makes it capable of

storing up to 4500 Fe3+ atoms, sequestering iron in a nontoxic form (39). The genes

encoding for H- and L- ferritin reside on different chromosomes and are expressed

independently. The H- subunit is capable of oxidizing Fe2+ to Fe3+, whereas the L-

subunit assists in hydrolysis and core formation. Ferritin functions as a buffer against

iron deficiency and overload, as it stores and releases iron in a controlled manner.

When iron levels are low, ferritin synthesis is decreased, less iron is stored and more

can be utilized by the cells; conversely, when iron levels are high, ferritin synthesis

increases, protecting cells from damage by excess iron (37). This regulation is mainly

through the well-characterized post-transcriptional, iron-dependent machinery based on

the interaction between IRPs and IREs. The IRE within ferritin is a stem loop structure

localized in the 5' UTR of both H- and L- ferritin mRNA. When iron is abundant, IRP1

assembles 4Fe-4S cluster and functions as cytosolic aconitase; during iron deficiency,

the cluster is disassembled and the protein binds IRE with high affinity (40). In the case

of ferritin, when IRP binds to its 5' UTR, translation is repressed (Figurel-3).

TF is a glycoprotein produced mainly in the liver with a molecular mass around 80

kDa. It contains two specific high-affinity ferric iron binding sites located in its amino-

and carboxy- terminus (1). The binding of TF to iron atoms is associated with binding of

an anion, usually bicarbonate (41). It has been established that TBI can be released by

acidification (42). Mammalian cells take up TBI through receptor-mediated endocytosis.

TFR1 is a homodimer of a 90 kDa glycosylated polypeptide with 61 amino acids N-

terminal of each subunit localized in the cytoplasm (43-45). This intracellular tail directs









IRE

kDa

LAMP1

LAMP2

LDL

LPS

LRP

mRNA

Fe-NTA

NTBI

PFA

RBC

RES

ROS

SFM

siRNA

Slc/SLC

Steap

TBI

TBS

TBST

Tf/TF

TfR1/TFR1

TfR2/TFR2

TM


Iron-responsive element

Kilodalton

Lysosomal-associated membrane protein 1

Lysosomal-associated membrane protein 2

Low density lipoprotein

Lipopolysaccharide

LDL receptor-related protein

Messenger RNA

Ferric nitrilotriacetic acid

Non-transferrin-bound iron

Paraformaldehyde

Red blood cell

Reticuloendothelial system

Reactive oxygen species

Serum-free medium

Small interfering RNA

Solute carrier

Six-transmembrane epithelial antigen of the prostate

Transferrin-bound iron

Tris-buffered saline

Tris-buffered saline Tween-20

Transferrin

Transferrin receptor 1

Transferrin receptor 2

Transmembrane









Until now, the only transmembrane transport protein implicated in TBI uptake by

hepatocytes was DMT1 (177). Data from the present study suggest that ZIP14, which is

expressed in at least 10 times greater amounts than is DMT1 in HepG2 cells, plays a

role in TBI uptake by hepatocytes.









cleavage site) and in the histidine-rich loop (immediately after the histidine 254 in the

"HHHGHNH" motif) of the pCMV-Sport6-mZipl4 vector (by primers P3+P4 or P5+P6,

Table 2-2). The3x Flag sequence was amplified from pCMV-3tag-3a with Spe I sites

added at each end (By primers P7+P8,Table 2-2), followed by lighting the PCR product

into the Spe I site-added vectors. All constructs were verified by DNA sequencing.

Primary Sequence Analysis and Bioinformatic Predictions

The mouse Zip14 gene sequence (NC_000080.5) and its protein sequences,

isoform a (NP_001128623.1) and isoform b (NP_659057.2), were obtained from the

NCBI website. Sequence alignment and consensus analysis were performed by using

Vector NTI 11.0 software (Invitrogen). Prediction of the transmembrane topology and

signal peptide sequence were performed by using the following online programs:

TMHMM (147), HMMTOP (148), TopPred (149), MINNOU (150), TmPred (151),

ConPred II (152), MEMSAT-SVM & MEMSAT III (153), Phobius (154), SOSUI (155),

SignalP3.0 & SignalP V2.0b2 (156), SPEPlip (157), PrediSi (158) and SIG-Pred

(bmbpcu36.leeds.ac.uk/prot_analysis/Signal.html). Prediction of protein glycosylation

status was done by using the online programs as follows: NetNGlyc 1.0

(http://www.cbs.dtu.dk/services/NetNGlyc/) and NetOGlyc 3.1 (159).

Construction of Asparagine Mutant Plasmids

Mouse Zipl4 contains five predicted N-linked glycosylation sites as asparagines

52, 75, 85, 100 and 455. The codons for each asparagine (N), were mutated to encode

for aspartic acid (D), giving different N-to-D mutants. Site-directed mutagenesis was

performed by using the Quickchange II Kit (Stratagene). In brief, 100 ng of double-

stranded DNA template (N-T-Flag-mZipl4 or C-T-Flag-mZipl4) was mixed with the

primers (forward and reverse primers, 125 ng each, Table 2-3), 10 mM dNTPs, 1 x









18. Gunshin, H., Mackenzie, B., Berger, U. V., Gunshin, Y., Romero, M. F., Boron,
W. F., Nussberger, S., Gollan, J. L., and Hediger, M. A. (1997) Nature 388, 482-
488

19. Tandy, S., Williams, M., Leggett, A., Lopez-Jimenez, M., Dedes, M., Ramesh, B.,
Srai, S. K., and Sharp, P. (2000) J. Biol. Chem. 275, 1023-1029

20. Hallberg, L. (1981) Annu. Rev. Nutr. 1, 123-147

21. Shayeghi, M., Latunde-Dada, G. O., Oakhill, J. S., Laftah, A. H., Takeuchi, K.,
Halliday, N., Khan, Y., Warley, A., McCann, F. E., Hider, R. C., Frazer, D. M.,
Anderson, G. J., Vulpe, C. D., Simpson, R. J., and McKie, A. T. (2005) Cell 122,
789-801

22. Laftah, A. H., Latunde-Dada, G. O., Fakih, S., Hider, R. C., Simpson, R. J., and
McKie, A. T. (2009) Br. J. Nutr. 101, 1150-1156

23. Qiu, A., Jansen, M., Sakaris, A., Min, S. H., Chattopadhyay, S., Tsai, E.,
Sandoval, C., Zhao, R., Akabas, M. H., and Goldman, I. D. (2006) Cell 127, 917-
928

24. Raffin, S. B., Woo, C. H., Roost, K. T., Price, D. C., and Schmid, R. (1974) J.
Clin. Invest. 54, 1344-1352

25. Sharp, P., and Srai, S. K. (2007) World J. Gastroenterol. 13, 4716-4724

26. Andrews, N. C., and Schmidt, P. J. (2007) Annu. Rev. Physiol. 69, 69-85

27. McKie, A. T., Marciani, P., Rolfs, A., Brennan, K., Wehr, K., Barrow, D., Miret, S.,
Bomford, A., Peters, T. J., Farzaneh, F., Hediger, M. A., Hentze, M. W., and
Simpson, R. J. (2000) Mol. Cell 5, 299-309

28. Vulpe, C. D., Kuo, Y. M., Murphy, T. L., Cowley, L., Askwith, C., Libina, N.,
Gitschier, J., and Anderson, G. J. (1999) Nat. Genet. 21, 195-199

29. McClung, J. P., and Karl, J. P. (2009) Nutr. Rev. 67, 100-104

30. Hentze, M. W., and Kuhn, L. C. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 8175-
8182

31. Frishman, D., and Hentze, M. W. (1996) Eur. J. Biochem. 239, 197-200

32. Guo, B., Phillips, J. D., Yu, Y., and Leibold, E. A. (1995) J. Biol. Chem. 270,
21645-21651


105









181. Wang, F., Kim, B. E., Petris, M. J., and Eide, D. J. (2004) J. Biol. Chem. 279,
51433-51441

182. Begum, N. A., Kobayashi, M., Moriwaki, Y., Matsumoto, M., Toyoshima, K., and
Seya, T. (2002) Genomics 80, 630-645

183. Girijashanker, K., He, L., Soleimani, M., Reed, J. M., Li, H., Liu, Z., Wang, B.,
Dalton, T. P., and Nebert, D. W. (2008) Mol. Pharmacol. 73, 1413-1423

184. Mellquist, J. L., Kasturi, L., Spitalnik, S. L., and Shakin-Eshleman, S. H. (1998)
Biochemistry 37, 6833-6837

185. Miletich, J. P., and Broze, G. J., Jr. (1990) J. Biol. Chem. 265, 11397-11404

186. Kornfeld, R., and Kornfeld, S. (1985) Annu. Rev. Biochem. 54, 631-664

187. Snider, M. D., and Robbins, P. W. (1982) J. Biol. Chem. 257, 6796-6801

188. Kato, Y., Sakaguchi, M., Mori, Y., Saito, K., Nakamura, T., Bakker, E. P., Sato,
Y., Goshima, S., and Uozumi, N. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 6488-
6493

189. Helenius, A., and Aebi, M. (2001) Science 291, 2364-2369

190. Nakamura, S., Takasaki, H., Kobayashi, K., and Kato, A. (1993) J. Biol. Chem.
268, 12706-12712

191. Helenius, A. (1994) Mol. Biol. Cell. 5, 253-265

192. Paulson, J. C. (1989) Trends Biochem. Sci. 14, 272-276

193. Helenius, A., Marquardt, T., and Braakman, I. (1992) Trends Cell Biol. 2, 227-231
194. Cuthbertson, J. M., Doyle, D. A., and Sansom, M. S. (2005) Protein Eng. Des.
Sel. 18, 295-308

195. Rogers, E. E., Eide, D. J., and Guerinot, M. L. (2000) Proceedings of the National
Academy of Sciences of the United States of America 97, 12356-12360

196. Mao, X., Kim, B. E., Wang, F., Eide, D. J., and Petris, M. J. (2007) J. Biol. Chem.
282, 6992-7000

197. Apweiler, R., Hermjakob, H., and Sharon, N. (1999) Biochimica et biophysica
acta 1473, 4-8

198. Ben-Dor, S., Esterman, N., Rubin, E., and Sharon, N. (2004) Glycobiology 14,
95-101


115









LIST OF TABLES


Table
page

2-1 List of Primers Used for PCR to Generate and Screen for Endogenous
ZIP14-3x Flag in HepG2 Cells ......... ............ ................... 61

2-2 List of Primers Used for Generating 3x Flag-tagged mZipl4 Vectors ................61

2-3 List of Primers Used for Generating Asparagine Mutants ..............................62

4-1 Bioinformatic Prediction of mZip14 Transmembrane (TM) Regions ..................98

4-2 Prediction of mZip14 Signal Peptide Cleavage Site............................................98












Non-TF bound iron


Receptor-
independent
route


Zipl4


DMT1


Labile iron pool

" Labile iron pool


Metabolic
processes


\Hepatocyte


Figure 1-7. Iron uptake by hepatocytes. Hepatocytes have several pathways for iron
uptake from the circulation including the uptake of TBI (Fe2-TF, diferric TF) via
TFR1, TFR2, and TFR-independent mechanisms. The TFR1 pathway is well
characterized. When diferric TF binds TFR1, the complex is internalized by
endocytosis, and iron is dissociated from TF during endosomal acidification.
Released iron is transferred to the cytoplasm for iron-related biological
functions or storage as ferritin. The resulting apoTF-TFR1 complex then
returns to the cell surface for reutilization. The detailed pathways through
TFR2 and the one independent of TFR1 and TFR2 remain to be elucidated,
but are considered to be important for iron uptake in hepatocytes. The uptake
of NTBI, which is present in the plasma during conditions of iron overload, is
likely mediated by Zipl4 and/or DMT1.


STFR1


IEndos

Endosome


. DMT1 o


i-e+


Diferric TF


r









Results

Structure of the murine Zip14 Gene and Sequence Alignment of its Two Protein
Isoforms

The murine slc39a14 gene (mZip14) consists of 10 exons spread over more than

40 kb on chromosome 14. It has two transcripts that encode two isoform proteins of

both 489 amino acids due to alternative splicing of the equal length of exon 5A or 5B

(Fig. 4-1A ). Sequence alignment indicates that the two isoforms share 95.9% identity

and 97.5% consensus positions (Fig. 4-1B). The high degree of conservation in amino

acids sequences suggests their common function. I studied mZipl4a, the isoform that

has been shown to transport iron (129).

Epitope Mapping of mZipl4

Analysis of the amino acid sequence of mZip14 by using different prediction

programs results in different sequences being identified as transmembrane helices

(Table 4-1). However, most programs predict a long N-terminus, a long extramembrane

loop which includes a histidine-rich region and a relatively short C- terminal tail.

Prediction programs can sometimes incorrectly assign the topology of proteins,

especially when the protein contains a signal peptide (154). Seven different programs

predict the presence of a signal peptide in mZip14 (Table 4-2). To investigate the

topology of mZip14 protein experimentally, immunofluorescence studies were

performed to examine the orientation of N-, C- termini and histidine-rich loop. HEK 293T

cells were transiently transfected with constructs encoding mZip14 with a 3x Flag

epitope at the N-terminus, C-terminus and the loop containing the histidine-rich region.

When immunofluorescence studies were performed to detect N-terminal-3xFlag

mZip14, a signal was clearly detectable at plasma membrane in the absence of









of DMT1 mutations in humans, severe liver iron loading occurred after blood transfusion

and administration of rEpo resulted in rapid improvement of anemia, suggesting the

existence of a DMT1-independent pathway for iron utilization by TBI in reticulocytes and

iron delivery by TBI and/or NTBI in the liver (93).

Surprisingly, overexpression of HFE reduced both TBI uptake and NTBI uptake in

HepG2 cells (136).The effect of HFE on NTBI uptake was abolished by ZIP14

knockdown in these cells, confirming the role of ZIP14 in NTBI uptake. Lack of

functional hereditary hemochromatosis protein, HFE, causes iron overload in the liver,

heart and pancreas, mostly in hepatocytes, which are the major sites for HFE

expression in the liver (167). The strongest evidence that suggests ZIP14 is involved in

TBI uptake is that overexpression of HFE decreased ZIP14 protein levels by promoting

ZIP14 degradation. The reduced ZIP14 levels were associated with diminished uptake

of not only NTBI but also TBI in HepG2 cells. However, whether or not ZIP14 is involved

in TBI uptake remains to be determined.

In this chapter, the function of ZIP14 in TBI uptake was studied in HEK 293T and

HepG2 cells. The subcellular localization of Zip14 was studied in HepG2 cells. By

overexpressing mZip14 in HEK293T cells, I not only found about 25% increased TBI

uptake, but also found that Zip14 can transport iron at pH 6.5, suggesting that ZIP14

could be a candidate for endosomal iron release. Importantly, siRNA-mediated

knockdown of endogenous ZIP14 decreased TBI uptake in HepG2 cells by 45%.

Consistent with a role in TBI uptake, ZIP14 was found to localize not only at the plasma

membrane, but also with endocytosed TF. Taken together, these results suggest that

ZIP14 may play a role in TBI uptake in hepatocytes.









isoform did not affect its late endosomal and lysosomal localizations, indicating that C-

terminal tail of DMT1(+IRE) does not include the targeting determinant.

DMT1 is not the Only Transporter Involved in Endosomal Iron Release

Iron is transported in the circulation between absorption, storage, and utilization

sites by TF (100). In human plasma, the TF concentration normally ranges from 22 to

35 pM and is about 20% to 50% occupied by iron (101). Iron taken up by means of

TFR1 must be transported across the endosomal membrane to be released into the

cytosol (depicted in Fig. 1-4). Based on the observation that the loss of TFR1 produces

more severe effects than the loss of DMT1, other transporters functioning in the TF

cycle have been hypothesized in animals (86). In b rats, which have a loss of function

mutation in DMT1, TBI uptake was still effective in crypt cells of small intestine, where

DMT1 was not detectable (102). Since crypt epithelial cells depend mainly on TF cycle

for iron uptake, this indicates that DMT1 is not required for the uptake of TBI by

duodenal crypt epithelial cells.

By inactivating the mouse DMT1 gene globally and in selected tissues through

gene targeting and homologous recombination, Gunshin et al. (86) found that fetal

DMT1 was not needed for maternofetal iron transfer but that DMT1 activity was

essential for intestinal non-heme iron absorption after birth. DMT1 was also required for

normal hemoglobin production during the development of erythroid precursors. Iron

dextran administered to DMT1'- mice markedly increased liver iron stores in both

hepatocytes and macrophages, indicating that hepatocytes and other cells must have

an alternative iron uptake mechanism.

A recent study has shown that expression of MCOLN1 (mucolipinl) which belongs

to the mucolipin subfamily of transient receptor potential (TRP) proteins, mediates iron









Discussion

ZIP14 mediates NTBI uptake into cells. A physiologic role for ZIP14 in NTBI

uptake is suggested by the observation that human ZIP14 is expressed most

abundantly in the liver, heart, and pancreas (132), the tissues that preferentially

accumulate iron in iron-overload disorders such as hereditary hemochromatosis.

In addition to rapidly clearing NTBI, the liver readily takes up TBI (162). Studies in

isolated primary mouse and rat hepatocytes suggest that NTBI and TBI uptake may

share a common iron transporter (141,160,164). A study in HepG2 cells suggested that

ZIP14 may represent this common transporter (136). Specifically, it has been found that

HFE expression resulted in downregulation of ZIP14, which was associated with

decreased uptake of not only NTBI, but also TBI. Up to now, DMT1 is the only known

iron transporter which can function in endosomal iron release. However DMT1 knockout

and its loss of function mutation animal models indicate that in hepatocytes and possibly

other cell types, there are other transporters which can function in this process

(74,75,86).

I tested the hypothesis that ZIP14 plays a role in the uptake of iron from TF,

specifically in endosomal iron release. Firstly, I tested if increased expression of Zip14

would affect TBI uptake. To do this, I used human HEK 293T cells, an easily

transfectable cell line. A novel finding form our study was that overexpressing Zip14

resulted in a 25% increased iron uptake from TF. Since DMT1 is the only known

transporter functioning in endosomal iron release, one would predict that

overexpression of DMT1 would enhance the uptake of TF-bound iron. However, Wetli et

al. (169) found no increase in TBI uptake in HEK 293T cells stably overexpressing

DMT1. Shindo et al. (138) also found no change in TBI uptake in HLF cells (a hepatoma









Kramer. Degrees of freedom were approximated using the Kenward-Rogers method

(DDFM = kr). A probability level of P < 0.05 was defined as a significant difference.









better efficiency (189). N-linked glycosylation has been proposed to have different

functions in various proteins, such as serving as a signal for intracellular sorting and

cell-cell interaction, helping protein folding and trafficking, promoting resistance to

proteases, preventing protein aggregation and maintaining the proper conformation

along with standard function of the protein (190,191). Removal of consensus

glycosylation sequences or inhibition of glycosylation often results in misfolding or

aggregation of proteins that fail to reach the functional states (189,192). The misfolded

proteins are often linked covalently to each other by disulfide bonds and their ultimate

fate is degradation (193). It is found that not all the potential asparagines sites can be

glycosylated. In fact, only one third of the potential Asn-X-Ser/Thr sites in proteins are

actually glycosylated and the efficiency of glycosylation depends on properly oriented

and accessible Asn-X-Ser/Thr sequon (186).

In this chapter, I investigated mZip14's membrane topology by both epitope

mapping and computer program prediction. I also examined its glycosylation status and

effect of N-linked glycosylation on its trafficking and iron transport activity. I found

mZipl4 has an extracellular N-terminus, the histidine-rich loop and the C-terminus are

both intracellular, suggesting there are seven transmembrane domains of this protein.

Mouse Zipl4 is glycosylated at asparagines 52, 75, 85 and 100, residues that are all in

the extracellular N- terminus. N-linked glycosylation does not affect plasma membrane

localization, but it is required for iron transport activity of mZipl4.









early development (55). FPN1 is also highly expressed in the intestinal basolateral

epithelium and in macrophages of the liver and spleen. FPN1 levels increase markedly

in macrophages after erythrophagocytosis (56). When FPN1 is inactivated in the

embryo proper, sparing the extraembryonic visceral endoderm, mice survive, but

become anemic shortly after birth when they must rely on intestinal absorption for iron

accumulation (55). Also, macrophages of the liver and spleen showed marked iron

retention under this circumstance, likely because of an inability to release recovered

iron from phagocytosed effete erythrocytes. These observations suggest that FPN1 is

the major iron exporter functioning in iron-recycling macrophages.

Normal adult plasma contains 3-4 mg of iron, almost all bound to TF. Most of

circulating iron is from RES recycling of RBC iron; only small amounts are contributed

by hepatic iron stores and intestinal absorption. But, erythrophagocytosis is not the only

way for macrophages to acquire iron. They can obtain iron from TF (57). Other

pathways in macrophages take up cytotoxic free heme and hemoglobin (Fig. 1-5).

Hemoglobin is the most abundant and functionally important protein in erythrocytes. But

once released from RBCs, it becomes highly toxic because of the oxidative properties of

heme (protoporphyrin IX and iron) via the Fenton reaction to produce ROS causing cell

injury or death (58).

Normally, only a small amount of extravascular hemolysis occurs due to

destruction of senescent erythrocytes, causing Hb release into plasma. Yet, under

various intravascular hemolysis-linked conditions, such as hemorrhage,

hemoglobinopathies, ischemia reperfusion, sickle cell disease or malaria, large amounts

of free Hb are released (58). Hemoglobin released into plasma is readily bound









stoichiometrically by the liver-derived plasma protein haptoglobin (Hp), forming a

hemoglobin-haptoglobin (Hb-Hp) complex (59). CD163, a hemoglobin scavenger

receptor present on the surface of monocytes and macrophages in the liver and several

other tissues, mediates the endocytosis and subsequent degradation of Hb-Hp complex

(60). The entire Hb-Hp complex is degraded in lysosomes to release heme and various

proteolytic products within the cells. The heme is then degraded by heme H01 to

release iron, carbon monoxide, and biliverdin. The free iron enters the same intracellular

pool as iron taken up from other sources. Mice in which the haptoglobin gene is

inactivated do not have obvious disorders in iron metabolism, suggesting that Hp does

not play a major role in normal iron metabolism (61).

Plasma Hb is quickly oxidized toferrihemoglobin when the binding capacity of Hp

is exceeded (58). Ferrihemoglobin dissociates into globin and ferriheme. Ferriheme is

transferred to hemopexin (Hx), forming Heme-Hx complex. This complex is internalized

by receptor-mediated endocytosis into macrophages, which express LDL receptor-

related protein (LRP)/CD91, the receptor for the heme-Hx complex (62). LRP-CD91-

mediated endocytosis is also a degradation pathway for all known ligands of this

receptor (63). In addition to macrophages, LRP/CD91 is highly expressed in several

other cell types including hepatocytes and neurons (62). Inactivation of the Hx gene in

mice does not lead to disturbed iron metabolism, suggesting that the role of Hx in iron

homeostasis is important only under pathological conditions (64). Hemopexin levels in

serum reflect how much heme is present in the blood. Therefore, a high Hx level

indicates that there has been significant degradation of heme-containing compounds

and low Hx levels are one of the diagnostic features of hemolytic anemia. Extracellular









100. Richardson, D. R., and Ponka, P. (1997) Biochim Biophys Acta 1331, 1-40

101. Chua, A. C., Graham, R. M., Trinder, D., and Olynyk, J. K. (2007) Crit. Rev. Clin.
Lab. Sci. 44, 413-459

102. Gates, P. S., Thomas, C., Freitas, E., Callow, M. J., and Morgan, E. H. (2000)
Am. J. Physiol. Gastrointest. Liver Physiol. 278, G930-936

103. Dong, X. P., Cheng, X., Mills, E., Delling, M., Wang, F., Kurz, T., and Xu, H.
(2008) Nature 455, 992-996

104. Bargal, R., Avidan, N., Ben-Asher, E., Olender, Z., Zeigler, M., Frumkin, A.,
Raas-Rothschild, A., Glusman, G., Lancet, D., and Bach, G. (2000) Nat. Genet.
26, 118-123

105. Chen, C. S., Bach, G., and Pagano, R. E. (1998) Proc. Natl. Acad. Sci. U. S. A.
95, 6373-6378

106. Kohgo, Y., Ikuta, K., Ohtake, T., Torimoto, Y., and Kato, J. (2008) Int. J. Hematol.
88, 7-15

107. Selden, C., Khalil, M., and Hodgson, H. J. (1999) Gut44, 443-446

108. Trinder, D., Zak, 0., and Aisen, P. (1996) Hepatology 23, 1512-1520

109. Chua, A. C., Herbison, C. E., Drake, S. F., Graham, R. M., Olynyk, J. K., and
Trinder, D. (2008) Hepatology 47, 1737-1744

110. Goldenberg, H., Seelos, C., Chatwani, S., Chegini, S., and Pumm, R. (1991)
Biochim Biophys Acta 1067, 145-152

111. Morgan, E. H. (1991) Comp Biochem. Physiol. A Comp. Physiol. 99, 91-95

112. Eisenstein, R. S. (2000) Annu. Rev. Nutr. 20, 627-662

113. Herbison, C. E., Thorstensen, K. D., Chua, A. C., Graham, R. M., Leedman, P.
J., Olynyk, J. K., and Trinder, D. P. (2009) Am. J. Physiol. Cell Physiol.

114. Sciot, R., Paterson, A. C., Van den Oord, J. J., and Desmet, V. J. (1987)
Hepatology 7, 831-837

115. Kawabata, H., Yang, R., Hirama, T., Vuong, P. T., Kawano, S., Gombart, A. F.,
and Koeffler, H. P. (1999) J. Biol. Chem. 274, 20826-20832


110









binding to the duodenal brush border membrane (20). Heme carrier protein 1 (HCP1)

was identified to be a heme import protein (21), but subsequent studies showed that it

was a folate transporter (22,23). The precise role of HCP1 in iron metabolism will

require further investigation. Following uptake into enterocytes, heme is broken down by

heme oxygenase 1 (HO); liberated ferrous iron enters the intracellular iron pool with the

iron absorbed from non-heme pathway (24).

Absorbed iron has two fates depending on the body's requirements. Some of the

iron extracted from the diet is stored in ferritin within the enterocytes, whereas some is

exported across the basolateral membrane of enterocytes. If body iron stores are

sufficient, and there is no increased erythropoietic demand, a large amount of newly

absorbed iron will be stored as ferritin in enterocytes. This intracellular iron is lost within

3 to 4 days as epithelial cells slough into the intestinal lumen (25,26).

The transfer of iron across the basolateral membrane of enterocytes into the

circulation is accomplished by the coordination of a transport protein ferroportinl

(FPN1) and a ferrioxidase, hephaestin (27,28). FPN1 is the only iron exporter identified

to date. Ferrous iron is exported across the basal membrane by FPN land then oxidized

by hephaestin before becoming bound to plasma TF.

Intestinal iron absorption is regulated by body iron stores, recent dietary iron intake

and erythropoietic activity. Two models have been proposed to explain how the

absorption of iron is regulated: the crypt programming model and the hepcidin model

(7,8). The first one is based on a post-transcriptional regulation scheme of iron

metabolism in which cellular iron levels are balanced mainly by the iron-responsive

element/iron regulatory protein (IRE/IRP) system (29).









TBS, antibody cross-reactivity was visualized by using enhanced chemiluminescence

(SuperSignal West Pico, Thermo Scientific) and x-ray film.

Subcellular Localization

Cells seeded on poly-L-lysine (Sigma) coated coverslips were washed two times

with PBS/' (PBS with 1mM MgCI2 and 0.1mM CaCI2) and fixed with 2%

paraformaldehyde (PFA) for 15 min at room temperature. Next, cells were washed three

times with PBS, permeabilized with 0.1% saponin for 10 min, and then washed three

times with PBS before blocking in 1% bovine serum albumin (BSA) for 30 min. For TF

labeling, cells were incubated with 30 [tg/ml Alexafluor 488-labeled human holo-TF

(Invitrogen) for 30 min at 37C and washed three times with PBS before fixation. To

stain the nuclei, cells were washed three times with PBS and incubated for 5 min with

10 pg/ml 4'-6-diamidino-2-phenylindole (DAPI). After several washes of PBS, coverslips

were mounted on microscope slides with mounting medium (Vector Laboratories) and

sealed with nail polish. Images were captured with a Leica TCS SP5 laser-scanning

confocal microscope.

Quantitative Determination of TF uptake

To determine if ZIP14 knockdown affected the uptake of TF, HepG2 cells were

incubated with 100 nM biotin-labeled holo-TF (Sigma). After 4 h, uptake and surface

binding of TF were stopped by trypsinization. After washing suspended cells twice in

cold SFM, cells were lysed in SDS lysis buffer. Cell extracts were analyzed by Western

blotting as described above, but with ImmunoPure HRP-conjugated streptavidin

(1:50,000 for 1 h) instead of antibodies.









Isolation of Plasma Membrane Proteins by Cell-surface Biotinylation

HepG2 cells were incubated overnight in SFM with 10 pM apo-TF or 10 pM holo-

TF. For wild-type mZip14 and the asparagine mutants, the transporters were expressed

in HEK293T cells by transient transfection as described above. Twenty-four hours after

transfection, the medium was removed, and the cells were washed twice with ice-cold

PBS. Plasma membrane proteins were isolated by using the Cell Surface Protein

Isolation Kit (Thermo Scientific) according to the manufacturer's protocol. Briefly, the

flasks were kept on ice, and all solutions were ice-cold for the rest of the procedure.

Each flask of cells was incubated with 10 ml of the membrane-impermeant biotinylation

reagent, NHS-SS-biotin (0.25mg/ml in PBS) for 30 min with very gentle shaking. After

biotinylation, each flask was added 500 pl quenching solution to quench the unreacted

NHS-SS-biotin. Cells were collected and lysed in 500 pl lysis buffer with 1 x protease

inhibitors (Roche) followed by centrifugation at 10,000 xg for 2 min at 4C. The clarified

supernatant was added to a spin column containing pre-washed immobilized

NeutrAvidin gel and incubated for 60 min at room temperature. After four washes,

biotinylated samples were incubated with 50 mM DTT in 1 x SDS-PAGE sample buffer

for 60 min at room temperature to cleave the disulfide bond and release biotinylated

proteins. Cell-surface expression of ZIP14, TFR2, Lamp1, tubulin, and Na, K ATPase

were determined by Western blotting as above, but with these additional antibodies:

mouse anti-Lampl or Na, K ATPase (1:1000, Santa Cruz) and mouse anti-tubulin,

clone B-5-1-2 (1:5000, Sigma).

RNA Isolation and Quantitative PCR (qPCR)

Total RNA was isolated from HepG2 and HEK 293T cells by using RNABee

(TelTest). Isolated RNA was treated with DNasel (Turbo DNA-free kit, Ambion) to









animals. Hfe'-Zip14'-, DMT1'-Zip14'- or Hfe'-DMT1-'Zip14'- double or triple knockout

animal will ideally be the better animal models.

The structure of a protein is always important for understanding its biological

function. The experimentally derived topology model of Zip14 from this study suggests

that the histidine-rich metal-binding motif is intracellular. Whether or not this motif is

involved in iron binding or intracellular iron level sensing, remains to be investigated.

Another ZIP protein signature motif "HEXPHEXGD" is also present in Zip14 with

the first histidine replaced with glutamic acid, resulting in the sequence "EEFPHELGD".

It has been shown that this motif is important for metal-binding and the glutamic acid

substitution does not affect its ability to transport zinc (132). The Zip14 topology model

from the present study suggests that this motif is within the third extracellular loop,

presenting the possibility that this motif is important for the recruitment and binding of

extracellular iron, thus important for the iron transport activity. However this hypothesis

needs to be tested.

The signal motif of Zip14 for endocytosis is another interesting aspect. If ZIP14

localizes in endosomes, a specific internalization signal may present in its primary

sequence. Endocytosis of cell surface protein is largely dependent on specific

internalization motifs located within the protein's cytoplasmic domains. For example the

tyrosine-containing motif with the consensus sequence of YXXP, where 0 represents a

hydrophobic residue, is present in LDL receptor (as NPVY) and in TFR1 (as YTRF)

(210,211), both proteins are well known to internalize via a receptor-mediated

endocytosis manner. Sequence analysis indicates that Zip14 has a conserved YSDI

motif in the cytoplasmic histidine-rich loop, which could serve as a potential


102









199. Rudd, P. M., Elliott, T., Cresswell, P., Wilson, I. A., and Dwek, R. A. (2001)
Science (New York, N. Y) 291, 2370-2376

200. Fan, G., Goldsmith, P. K., Collins, R., Dunn, C. K., Krapcho, K. J., Rogers, K. V.,
and Spiegel, A. M. (1997) Endocrinology 138, 1916-1922

201. Chen, Q., Miller, L. J., and Dong, M. (2010) Am. J. Physiol. Endocrinol. Metab.
299, E62-E68

202. Nguyen, T. T., and Amara, S. G. (1996) J. Neurochem. 67, 645-655

203. Liu, X., Davis, D., and Segaloff, D. L. (1993) J. Biol. Chem. 268, 1513-1516

204. Mohrmann, K., van Eijndhoven, M. A., Schinkel, A. H., and Schellens, J. H.
(2005) Cancer Chemother. Pharmacol. 56, 344-350

205. Soroka, C. J., Xu, S., Mennone, A., Lam, P., and Boyer, J. L. (2008) BMC Cell
Biol. 9, 57

206. Gong, Q., Anderson, C. L., January, C. T., and Zhou, Z. (2002) Am. J. Physiol.
Heart Circ. Physiol. 283, H77-84

207. Nilsson, I. M., and von Heijne, G. (1993) J. Biol. Chem. 268, 5798-5801

208. Guan, C., Ye, C., Yang, X., and Gao, J. (2010) Genesis 48, 73-85

209. Brunt, E. M. (2005) Semin. Liver Dis. 25, 392-401

210. Girones, N., Alverez, E., Seth, A., Lin, I. M., Latour, D. A., and Davis, R. J. (1991)
J. Biol. Chem. 266, 19006-19012

211. Chen, W. J., Goldstein, J. L., and Brown, M. S. (1990) J. Biol. Chem. 265, 3116-
3123


116













NC siRNA: +
ZIP14siRNA: -


- + +


anti-Flag
(ZIP14)


- 50 KDa


antiTFR1 75 KDa


anti-UTR2 -1,, KDa

50KDa
anti-Actin


NC siRNA: + + -
ZIP14 siRNA: + +
TF-Biotin: -
anti-Flag I
(ZIP14) .


Avidin-HRP
(TF)icn

anti-Actin


(.-
- 800-
d o-

oT 600-
E
E 400-

200-


F
P










iRNA c -ntrl
sir NA control


'<0.05


1


r


siRNA ZIP14


+ + -
- + +
+ + + +


a E -


-50 KDa


75 KDa
0 -50 KDa
ago&. o-Ow4MMM


Figure 3-7. Knockdown of ZIP14 decreases TBI uptake by HepG2 cells. A) Knockdown
of ZIP14 in HepG2 cells does not affect the expression of TFR1 or TFR2.
HepG2 cells were transfected with negative control (NC) or ZIP14 siRNA for
72 h prior to Western blot analysis for ZIP14, TFR1, TFR2, and actin as a
lane loading control. B) Knockdown of ZIP14 decreases the uptake of TBI.
HepG2 cells transfected with NC or ZIP14 siRNA for 72 h were incubated with
100 nM 59Fe-TF for 4 h. Cells were harvested and cell-associated radioactivity
was determined by y-counting. The amount of TBI taken up by cells is
expressed as cpm per mg of protein. Data represent the mean SEM of
three independent experiments. C) Knockdown of ZIP14 does not affect the
uptake of TF. HepG2 cells transfected with NC or ZIP14 siRNA for 72 h were
incubated with or without 100 nM biotin-labeled holo-TF. After 4 h, cell lysates
were analyzed by Western blotting for ZIP14, TF, and actin as a lane loading
control. The results shown are representative of one of three experiments
without significant variation between experiments.





0(I









release from late endosomes and lysosomes in HEK 293T cells (103). Bargal et al.

(104) identified the MCOLN1 gene, which encodes a 580-amino acid protein termed

mucolipinl. The MCOLN1 protein contains 1 transmembrane helix in the N-terminal

region and at least 5 transmembrane domains in the C-terminal region. Mutations in the

human MCOLN1 gene cause Mucolipidosis type IV disease which is an autosomal

recessive neurodegenerative lysosomal storage disorder characterized by psychomotor

retardation, ophthalmologic abnormalities, as well as iron-deficiency anemia (105).

In summary, DMT1 can function in both dietary iron absorption and endosomal

iron release. However, DMTI- mice were born alive with elevated hepatic stores,

indicating that hepatocytes have an alternative DMT1-independent iron uptake pathway

and that DMT1 is not required for maternofetal iron transport, which relies primarily on

TF-bound iron delivery. Moreover, in Belgrade rats, DMT1 is not required for TBI uptake

of small intestine crypt epithelial cells. Human cases of DMT1 mutations also indicate

the existence of DMT1-independent iron uptake by both reticulocytes and hepatocytes.

Therefore, other endosomal transporters must exist for transporting iron from TF into

the cytosol.

Iron Uptake by Hepatocytes

The liver plays at least three major roles in iron homeostasis. Firstly, it is the main

storage organ for iron. Hepatocytes as well as Kupffer cells serve as iron depots, in

which excess iron is stored as ferritin and hemosiderin (106). Secondly, the liver

regulates iron transfer into and around the body through the peptide hormone hepcidin

production. Thirdly, the liver is the major site for the synthesis of iron-related plasma

proteins such as TF and ceruloplasmin (Cp) (65). Hepatocytes comprise about 80% of









(71). The tissue distribution profile indicates that the exon 1B isoform is ubiquitously

expressed, whereas the expression of exon 1A isoform is tissue-specific and particularly

abundant in the duodenum and the kidney. The same study also found that, in mice,

DMT1 regulation in the kidney is associated with the presence of an IRE in the 3' UTR,

whereas in the duodenum, iron regulation is most strongly associated with the presence

of exon 1A. The multiple isoforms of DMT1 clearly add to the complexity for studying the

expression, function, and regulation of this protein.

Role of DMT1 in Iron Absorption

In rodent models, such as Belgrade (b) rats and mk mice, mutation of a single

nucleotide in the DMT1 gene results in a substitution of arginine for glycine at position

185 (G185R), leading to anemia and iron deficiency (74,75). The iron deficiency results

from impaired intestinal iron absorption and erythroid iron utilization. The phenotype of b

rats was first described in 1966, when inherited anemia was reported among the

offspring of X ray-irradiated rats in a nuclear science laboratory in Belgrade in ex-

Yugoslavia (76). The Belgrade (b) laboratory rat suffers from anemia accompanied by

elevated plasma iron and iron-binding capacity, decreased stainable iron in tissues and

decreased growth rate. Oats et al. (77) showed that the reduced uptake of both ferric

and ferrous iron in homozygous b rats probably involves a defective iron carrier

associated with the apical membrane of the duodenum.

Using a positional cloning approach, Fleming et al. (74,75) identified DMT1 as the

defective gene for both b rats and mk mice. Injection of RNA synthesized from DMT1

cDNA in Xenopus oocytes promoted the uptake of iron as well as other divalent cations,

including manganese, cobalt, and zinc (18). Uptake of metals by DMT1 is pH-

dependent, involving proton symport. In Caco2 cells, DMT1 was shown to transport iron









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

FUNCTIONAL AND STRUCTURAL ANALYSIS OF THE METAL-ION
TRANSPORTER ZIP14


By

Ningning Zhao

August 2010

Chair: Mitchell D. Knutson
Major: Nutritional Sciences

ZIP14 solutee carrier family 39, member 14, SLC39A14) is a transmembrane

metal-ion transporter that is abundantly expressed in the liver, heart, and pancreas.

Previous studies of HEK 293 cells and the hepatocyte cell lines AML12 and HepG2

established that ZIP14 mediates the uptake of non-transferrin-bound iron (NTBI), a form

of iron that appears in the plasma during iron overload disorders. I investigated the role

of ZIP14 in transferrin-bound iron (TBI) uptake and determined the subcellular

localization of ZIP14 in HepG2 cells. It was found that overexpression of mouse Zip14

(mZip14) in HEK 293T cells increased the uptake of TBI without increasing levels of

transferring receptor 1 (TFR1). I also found that mZipl4 can mediate the transport of iron

at pH 6.5, the pH at which iron atoms dissociate from transferring (TF) within endosomes.

I used an adeno-associated viral (AAV) system to generate a HepG2 cell line

expressing a Flag-tagged ZIP14 allele, allowing specific and sensitive detection of

endogenous ZIP14 in these cells. Confocal microscopic analysis detected ZIP14 at the

plasma membrane and in TF-containing endosomes. Knockdown of endogenous ZIP14

with siRNA did not decrease TFR1 or TFR2 levels, but resulted in a 45% reduction in









remove any contaminating genomic DNA. First-strand cDNA was synthesized from the

isolated RNA by using the High-Capacity cDNA Archive kit (Applied Biosystems).

Quantitative RT-PCR was performed using iQ SYBRGreen Supermix (Bio-Rad) and an

Applied Biosystems 7300 realtime PCR system. Copy numbers of DMT1 and ZIP14

mRNA were calculated by comparing Ct values obtained from HEK 293T and HepG2

RNA to those obtained from standard curves generated by using the plasmids

pBluescriptR-human DMT1 (BC100014, Addgene) and pXL4-human ZIP14

(NM_001135153.1, Open Biosystems). The primers used for ZIP14 (forward: 5'-

CTGGACCACATGATTCCTCAG-3'; reverse: 5'-GAGTAGCGGACACCTTTCAG-3') and

DMT1 (5'-TGGTTCTGACTCGCTCTATTGC-3'; reverse: 5'-

CATTCATCCCTGTTAGATGCTCTACA-3') were designed to target all known variants

of ZIP14 and DMT1 mRNA.

Construction of GFP Fusion and Flag-tagged mZipl4 Plasmids

An expression vector for C-terminal mZip14-EGFP was constructed by amplifying

mouse Zipl4 coding sequence (BC021530) from pCMV-Sport6-mZip14 vector (Open

Biosystems) and lighting into pEGFP-N1 (Kindly provided by Dr. Ivana DeDomenico,

University of Utah). To make a C-terminal mZip14-3x Flag vector (C-T-Flag-mZip14),

the 3x Flag sequence followed by a stop codon was amplified from pCMV-3tag-3a

(Stratagene) by a forward primer linked with a Sal I site and a reverse primer linked with

Not I site (P1+P2, Table 2-2). The PCR product was ligated into mZip14-EGFP vector

after excision of the EGFP sequence by Sal I and Not I restriction enzymes. To insert

the 3x Flag sequence into the N-terminus and the histidine-rich region loop (N-T-Flag-

mZip14 and His-Flag-mZipl4), site-directed mutagenesis (method see below) was first

used to add an Spe I restriction site in the N-terminus (after the predicted signal peptide









LIST OF REFERENCES

1. Crichton, R. R., and Charloteaux-Wauters, M. (1987) Eur. J. Biochem. 164, 485-
506

2. Hentze, M. W., Muckenthaler, M. U., and Andrews, N. C. (2004) Cell 117, 285-
297

3. Aruoma, O. I., Halliwell, B., Laughton, M. J., Quinlan, G. J., and Gutteridge, J. M.
(1989) Biochem J 258, 617-620

4. Andrews, N. C. (1999) N. Engl. J. Med. 341, 1986-1995

5. Smith, R. S. (1965) Arch. Dis. Child. 40, 343-363

6. Andrews, N. C. (2000) Nat. Rev. Genet. 1, 208-217

7. Siah, C. W., Ombiga, J., Adams, L. A., Trinder, D., and Olynyk, J. K. (2006) Clin.
Biochem. Rev. 27, 5-16

8. Munoz, M., Villar, I., and Garcia-Erce, J. A. (2009) World J. Gastroenterol. 15,
4617-4626

9. Andrews, N. C. (2008) Blood 112, 219-230

10. Sharp, P., and Srai, S. K. (2007) World J Gastroenterol 13, 4716-4724

11. McKie, A. T., Barrow, D., Latunde-Dada, G. O., Rolfs, A., Sager, G., Mudaly, E.,
Mudaly, M., Richardson, C., Barlow, D., Bomford, A., Peters, T. J., Raja, K. B.,
Shirali, S., Hediger, M. A., Farzaneh, F., and Simpson, R. J. (2001) Science 291,
1755-1759

12. Gunshin, H., Starr, C. N., Direnzo, C., Fleming, M. D., Jin, J., Greer, E. L.,
Sellers, V. M., Galica, S. M., and Andrews, N. C. (2005) Blood 106, 2879-2883

13. Frazer, D. M., Wilkins, S. J., Vulpe, C. D., and Anderson, G. J. (2005) Blood 106,
4413; author reply 4414

14. Ohgami, R. S., Campagna, D. R., McDonald, A., and Fleming, M. D. (2006)
Blood 108, 1388-1394

15. Han, O., Failla, M. L., Hill, A. D., Morris, E. R., and Smith, J. C., Jr. (1995) J.
Nutr. 125, 1291-1299

16. Glahn, R. P., and Van Campen, D. R. (1997) J. Nutr. 127, 642-647

17. Swain, J. H., Tabatabai, L. B., and Reddy, M. B. (2002) J. Nutr. 132, 245-251


104









are glycosylated. It has been shown that minimum distance for

oligosaccharyltransferase required for N-linked glycosylation active site is about 12-14

amino acid residues above the membrane and is oriented nearly parallel to the

membrane surface (207). In the present study, the gel electrophoresis data indicated

that all four potential asparagine sites at amino- terminus are linked with glycan,

consistent with the ecto-orientation of the amino- terminus. I further showed that

mutation of all four N-linked glycosylation sites does not block the plasma membrane

trafficking of mZipl4 protein, but removing the glycosylation of mZipl4 has functional

consequences. It has been demonstrated here that the iron transport activity of mZipl4

decreased significantly without N-linked glycan.

Data presented here provide for the first time experimental support for the

proposed seven-TM domain model of Zipl4, and provide evidence that N-linked

glycosylation is not required for membrane localization but it is important for iron

transport activity of Zipl4.









Table 2-1. List of Primers Used for PCR to Generate and Screen for Endogenous
ZIP14-3x Flag in HepG2 Cells


Primer Set
Knock-in Primers
Left arm

Right arm

Cre-specific
primers
Screening
primers


Forward Primer

5' GGGAAAGUGCAACCTT
GAACTCCGGAGC 3'
5'- GGTCCCAUGGCTCTGC
CAAGAGCCTG- 3'
5' ATATTGCGGCCGCAAGC
TTGGCCCATTGCATAC 3'
P1: 5'- CCATTCAGCGGTTTTT
AAGGGGGC- 3'
P3: 5' GGATTCATCGACTGT
GGCCG-3'
P5: 5' GGCCTCCTGACTGGA
TTCACC- 3'


Reverse Primer

5' GGAGACAUTCCCCAATC
TGGATCTGTCCTGAAT 3'
5'- GGCATAGUGCACCCCAT
TTCTACAAGTCAGC 3'
5' AATAAGCGGCCGCCGCG
TTAATGGCTAATCGCC 3'
P2: 5' GTTGTGCCCAGTCAT
AGCCG- 3'
P4: 5' CAAGGGCTCCACAGT
GGCTAAG- 3'
P6: 5' GAGACTGGTTACCAG
GGCAGC -3'


Table 2-2. List of Primers Used for Generating 3x Flag-tagged mZipl4 Vectors

Primer Name Primer Sequence


3x Flag
al I + Not I)
N-T Spe I

His Spe I


3x Flag Spe I


Forward P1: 5' ATCGATACCGTCGACCTCGAG 3'
Reverse P2: 5' AATAAGCGGCCGCCTATTTATCGTCATCAT 3'
Forward P3: 5' CTGCCGCCCCTCACTAGTGCCACCTCC 3'
Reverse P4: 5' GGAGGTGGCACTAGTGAGGGGCGGCAG 3'
Forward P5: 5' ATCACGGGCATAACCATACTAGTTTTACCTCCG
AGACACT- 3'
Reverse P6: 5' AGTGTCTCGGAGGTAAAACTAGTATGGTTATGC
CCGTGAT- 3'
Forward P7: 5'- ATATTACTAGTCAGATTACAAGGATGACGACGA
TA- 3'
Reverse P8: 5' AATAAACTAGTTTTATCGTCATCATCTTTGTAGT
CC -3'


(s










I I Control
M Zipl4


P<0.05
7--


6.5
pH


P>0.05






5.5


Figure 3-2. pH dependence of Zipl4-mediated iron transport. HEK 293T cells
transfected with pCMVSport2 (control) or pCMVSport6/mouse Zip14 were
incubated with 2 pM 59Fe-ferric citrate for 1 h in uptake buffer at pH 7.5, 6.5,
and 5.5. The amount of 59Fe taken up by cells is expressed as cpm per mg of
protein. Data represent the mean SEM of three independent experiments.









When recombinant DMT1(-IRE) and DMT1(+IRE) were expressed simultaneously

in HEp2 cells (a human larynx carcinoma cell line, which highly express DMT1(+IRE)

isoform), the two isoforms did not completely colocalize with each other, indicating

distinct subcellular compartmentalization (98). DMT1 (+IRE) localized in late

endosomes and lysosomes in transfected HEp-2, HeLa (a human cervical cancer cell

line) and COS-7 cells. In HEp-2 cells, the early endosomal markers EEA1 (early

endosomal antigen 1) and TFR1 were not colocalized with DMT1, but most of the

intracellular puncta of DMT1 completely colocalized with LAMP2 (lysosome-associated

membrane protein 2). Subcellular fractionation by sucrose gradient revealed that the

DMT1(+IRE) isoform co-sedimented with the late endosomal and lysosomal membrane

proteins LAMP1 (lysosome-associated membrane protein 1) and LAMP-2, but not with

the TFR1 in early endosomes (99). The acidic environment in the late endosomal and

lysosomal may provide DMT1 the proton gradient required for transport activity. DMT1(-

IRE) colocalized with the early endosomal markers TFR1 and EEA1 (98), indicating its

localization in early endosomes.

In DMT1, the carboxy cytoplasmic tail is the only different part of +IRE and -IRE

forms, indicating the possible existence of targeting signals for different localization in

this domain. Substitution of the Tyr555 or Leu557 with Ala significantly affected the early

endosomal localization of DMT1(-IRE) isoform, resulting in its mistargeting to late

endosomes and lysosomes. It was concluded that the Y55XLXX sequence in the C-

terminal cytoplasmic tail of DMT1(-IRE) is critical for the early endosomal targeting (98).

Sequential deletion of almost all the amino acids in the C-terminal tail of DMT1(+IRE)








Nonpermeabilized
Sawonin


Permeabilized
+ Saponin


Zipl4-GFP





B


anti-GFP





C


Merge





Figure 4-3. The carboxy terminus of mZipl4 is intracellular.HEK 293T cells were
transiently transfected with mZipl4-GFP vector (GFP at C-terminus of
mZipl4). A) Green fluorescence signal of Zipl4-GFP can be seen under both
permeabilized and nonpermeabilized conditions. B) Zipl4-GFP can be
detected by GFP antibody only after saponin treatment. C) Merge picture of
panels A-B. Data represent two independent experiments.










STOP

WT allele ZIP4
->0.26 Kbp<-
P5 P6


STOP

4 ,p


Flag + NeoR allele ZIP4

> 1.8 Kbp




Flag allele ZIP14
P- 0
P5


NEOR -

P2 P Kb P4
P2 P3 1.6 Kbp
P2 P3 P4


STOP


1-o----
59 Kbp <
P6


WT WT/Flaa


WT
NC siRNA:
ZIP14 siRNA:


anti-Flag
(ZIP14)



anti-Actin I _


WT/Flag
- + -
- +


50 KDa

50 KDa


Figure 3-5. Targeted knock-in of 3x Flag into the ZIP14 locus of HepG2 cells. A)
Diagrams of the wild-type (WT) ZIP14 allele, the Flag + NeoR allele after
homologous recombination with the targeting vector, and the Flag allele after
excising the neomycin cassette with Cre recombinase. B) PCR of genomic
DNA identifies clones with the Flag + NeoR allele and Flag allele. C) Western
blot analysis of HepG2 cells expressing Flag-tagged ZIP14. Knockdown of
endogenous ZIP14 in HepG2 cells using siRNA targeting ZIP14. To indicate
protein loading among lanes, blot was stripped and reprobed for actin.


P1 + P;



P3 + P-


P5 + P6


1.5 Kbp


1.5 Kbp


0.5 Kbp


0.2 Kbp









epithelium. Cytochrome b-like ferrireductase (Dcytb) is thought to be the responsible

enzyme (11). However, by Dcytb gene knockout in mice, Gunshin et al. (12)

demonstrated that the absence of Dcytb did not impair accumulation of body iron stores

when mice were fed normal chow diet, indicating that there was no major effect on

intestinal absorption. It was concluded that Dcytb is dispensable for intestinal iron

absorption in mice. However, the role of Dcytb in intestinal iron absorption remains

unclear as direct iron absorption was not measured in the knockout study (13).

Moreover, mice are capable of synthesizing ascorbic acid and may have less need for a

duodenal surface ferric reductase (10). The recently identified six-transmembrane

epithelial antigen of the prostate 2 (Steap2) protein might be another candidate for this

ferrireductase role (14). There are also various dietary components, such as ascorbic

acid, cysteine and histidine, which are capable of reducing Fe3+ to Fe2+ (15-17).

After being reduced by Dcytb or other reducing agents, Fe2+ becomes a substrate

for a transmembrane transporter. Gunshin et al. (18) identified in rats a metal-ion

transporter, named DCT1 divalentt cation transporter 1, also called DMT1, divalent

metal transporter 1) and found that this protein, which is upregulated by dietary iron

deficiency, may represent a key mediator of intestinal iron absorption. In Caco2 cells

(human epithelial colorectal adenocarcinoma cells), it has been shown that DMT1 is

expressed in the apical membrane (19).

Non-vegetarian diets contain another form of iron, heme iron (primarily from

hemoglobin and myoglobin), which may account for 10% of the dietary iron intake.

Because heme iron is more bioavailable than non-heme iron, it may contribute half of

the total iron absorbed in a western meat-rich diet (20). Absorption of heme occurs by









Several studies have reported cases of DMT1 mutations in humans. Mims et al.

(91) first reported a case of a female with severe hypochromic microcytic anemia and

iron overload. They found a G-to-C mutation in exon 12 (DMT1 1285 G-C), resulting in

a conservative glutamic acid to aspartic acid (E399D) single amino acid substitution.

However, the predominant effect of this mutation is exon 12 skipping during processing

of the mRNA present in erythroid cells. Removal of exon 12 deletes transmembrane

domain 8, which may interfere with proper protein insertion into the membrane. It has

been shown that E399D resides in the 4th predicted intracellular loop of DMT1 and

forms part of a highly conserved transport signature motif among species (92). By

expressing mutant E399D, E399Q, E399A in LLC-PK1 kidney cells (epithelial cell line

derived from porcine kidneys), it has been found that the mutants are fully functional in

term of stability and targeting to the membrane, and are also transport-competent,

indicating that DMTIG1285C is not a complete loss of function. In fact, G1285C mutation

greatly increases the exon 12 skipping risk (90% compared with 10% in healthy

individuals and 50% heterozygote phenotypically normal sibling). The limited amount of

functional E399D produced in the DMT1G1285C patient may be adequate for iron

absorption, but not sufficient for efficient utilization in the erythrocytes. Thus, the clinical

phenotype appears to be primarily due to exon 12 skipping rather than one amino acid

substitution. Interestingly, liver biopsy indicated that this patient had severe iron loading

in both hepatocytes and Kupffer cells, suggesting the existence of another iron transport

pathway functioning in the liver.

lolascon et al. (93) reported an infant with hypochromic, microcytic anemia, and

hepatic iron overload with increased serum iron, TF saturation, and serum ferritin levels









Role of the IRE/IRP regulatory system in cellular iron homeostasis: Cellular

iron homeostasis is attained by coordinated and balanced expression of proteins

involved in iron uptake, export, storage and usage. The posttranscriptional regulation

mediated by the IRE/IRP system appears to be essential, which is a specific mRNA-

protein interaction in the cytoplasm (2). An overview of IRE/IRP regulatory network is

illustrated in Figurel-3. In this system, particular cis-regulatory elements, stem loop

structures called iron-responsive elements (IREs) in the untranslated regions (UTR) of

respective mRNAs, are recognized by trans-acting proteins, known as iron-regulatory

proteins (IRPs), which are cytosolic iron sensors that control the rate of mRNA

translation or stability (30). Two IRPs (IRP1 and IRP2) have been identified. Both IRPs

belong to the aconitase family proteins and share 64% amino acid sequence identity,

but different mechanisms control their activities in response to cellular iron levels (31).

During iron-deficient conditions, IRP1 binds IREs. Binding at the 5' UTR blocks

ribosome access to mRNA, inhibiting translation, whereas binding at the 3' UTR leads

to increased mRNA stability by preventing mRNA degradation. Under iron-sufficient

conditions, an iron-sulfur cluster (4Fe-4S) assembles in IRP1, preventing its binding to

IREs. During iron-deficiency, IRP1 is open and accessible for interaction with IREs. In

contrast, IRP2 does not contain the iron-sulfur cluster and regulation by iron is through

proteasomal degradation (32,33).

The crypt programming model: This model proposes that the intracellular iron

level of the crypt cells corresponds to the body's iron store level, which in turn regulates

dietary iron absorption via mature villus enterocytes (7,8). Iron uptake from plasma

through TBI into crypt cells determines the iron content within the cells. The intracellular









Detection of Zipl4-GFP in recycling endosomes further implicates this protein in the

assimilation of iron from TF.

ZIP14 and DMT1 mRNA Copy Numbers in HepG2 and HEK 293T Cells

ZIP14 is most abundantly expressed in the liver (131,132), a major organ of iron

metabolism. It has been reported that ZIP14 mRNA levels (relative to GAPDH mRNA)

were approximately 6-fold higher than DMT1 mRNA levels in HepG2 cells (136). Here I

measured absolute mRNA copy numbers of ZIP14 in HepG2 and HEK 293T cells, and

compared them to DMT1 mRNA copy numbers. In HepG2 cells, ZIP14 mRNA copy

numbers were found to be 11-fold greater than DMT1 copy numbers (Fig. 3-4).

Moreover, ZIP14 transcript levels in HepG2 cells were found to be 11-fold greater than

in HEK 293T cells. Likewise, in human tissues ZIP14 is much more abundantly

expressed in the liver than in the kidney (132). The high expression level of ZIP14 in the

liver and HepG2 cells suggests that ZIP14 plays an important role in hepatocytes.

Epitope tagging of Human Endogenous ZIP14 in HepG2 Cells

To investigate the role and subcellular localization of ZIP14 in HepG2 cells, I first

generated a cell line that expresses Flag-tagged ZIP14 from its endogenous locus. The

addition of a Flag-tag to ZIP14 allows for highly specific and sensitive detection of the

endogenous protein by the mouse anti-Flag M2 monoclonal antibody. Flag-tag-

encoding DNA was knocked into the endogenous ZIP14 locus by homologous

recombination (Fig. 3-5A). Correctly targeted neomycin-resistant (NeoR) clones were

identified by genomic PCR. Primers used for the knock-in and screening are shown in

Table 2-1. Screening primers were designed to regions of the Flag + NeoR allele

predicted to be ~1.8 Kbp (P1 + P2) and ~1.6 Kbp (P3 + P4). Fig. 4B shows the PCR

identification of a correctly targeted clone. DNA sequencing of the PCR products









84. Muir, A., and Hopfer, U. (1985) Am. J. Physiol. Gastrointest. Liver Physiol. 248,
G376-379

85. Griffiths, W. J., Kelly, A. L., Smith, S. J., and Cox, T. M. (2000) QJ Med. 93, 575-
587

86. Gunshin, H., Fujiwara, Y., Custodio, A. O., Direnzo, C., Robine, S., and Andrews,
N. C. (2005) J. Clin. Invest. 115, 1258-1266

87. Edwards, J. A., Sullivan, A. L., and Hoke, J. E. (1980) Blood 55, 645-648

88. Bowen, B. J., and Morgan, E. H. (1987) Blood 70, 38-44

89. Garrick, M. D., Gniecko, K., Liu, Y., Cohan, D. S., and Garrick, L. M. (1993) J.
Biol. Chem. 268, 14867-14874

90. Garrick, L. M., Dolan, K. G., Romano, M. A., and Garrick, M. D. (1999) J. Cell
Physiol. 178, 349-358

91. Mims, M. P., Guan, Y., Pospisilova, D., Priwitzerova, M., Indrak, K., Ponka, P.,
Divoky, V., and Prchal, J. T. (2005) Blood 105, 1337-1342

92. Lam-Yuk-Tseung, S., Mathieu, M., and Gros, P. (2005) Blood Cells Mol. Dis. 35,
212-216

93. lolascon, A., d'Apolito, M., Servedio, V., Cimmino, F., Piga, A., and Camaschella,
C. (2006) Blood 107, 349-354

94. Beaumont, C., Delaunay, J., Hetet, G., Grandchamp, B., de Montalembert, M.,
and Tchernia, G. (2006) Blood 107, 4168-4170

95. Canonne-Hergaux, F., Zhang, A. S., Ponka, P., and Gros, P. (2001) Blood 98,
3823-3830

96. Picard, V., Govoni, G., Jabado, N., and Gros, P. (2000) J. Biol. Chem. 275,
35738-35745

97. Gruenheid, S., Canonne-Hergaux, F., Gauthier, S., Hackam, D. J., Grinstein, S.,
and Gros, P. (1999) J. Exp. Med. 189, 831-841

98. Tabuchi, M., Tanaka, N., Nishida-Kitayama, J., Ohno, H., and Kishi, F. (2002)
Mol. Biol. Cell 13, 4371-4387

99. Tabuchi, M., Yoshimori, T., Yamaguchi, K., Yoshida, T., and Kishi, F. (2000) J.
Biol. Chem. 275, 22220-22228


109









shares 69% similarity with hZIP14, was predicted to have seven transmembrane

domains with an extracellular N-terminus and intracellular C-terminus (182). In the same

study, hZIP14 was also predicted to have seven transmembrane domains. There are

several other studies which reported the predicted topology models for ZIP14 protein.

Three of them predicted ZIP14 contains eight transmembrane domains, among which

one predicted hZIP14 topology (132), two predicted mouse Zip14 (mZip14, which

shares 83% identity with hZIP14) (134,183). One study also predicted that both amino-

and carboxy- termini of mZip14 were located extracellularly and the histidine-rich

region-containing loop localized extracellularly as well (129). None of these models,

however, has been experimentally investigated.

Recombinant human ZIP14 is sensitive to PNGase F digestion, indicating that it is

an N-linked glycoprotein (132). N-linked glycosylation [Glycosylation at the asparagine

(Asn or N) residue] is an important modification of protein, which starts co-translationally

with the transfer of the pre-synthesized oligosaccharide chain from a lipid precursor to

an asparagine residue of the nascent protein in the sequence Asn-X-Ser/Thr (in some

case, it can be Asn-X-Cys), where X is any amino acid except proline (184,185). After

the initial glycosylation, the processing of the carbohydrate side chain takes place

primarily in endoplasmic reticulum (ER) and the Golgi apparatus by the sequential

addition and removal of monosaccharide units (186). Glycosylation on asparagines of

membrane protein requires that potential asparagine is in the ER lumen during

glycoprotein maturation, and is extracellular after protein incorporation into the plasma

membrane (187,188). Addition of carbohydrate side chains gives proteins branched and

mobile polar domains, which helps cells secrete proteins of greater complexity and with












M 1 2 2 4 SA SB 6 7 8 9 10




Iofom Isofrmb b



Intron

SExon : UTR CDS




(1) ,10 20 i30 ,40 55
B mZipl4a (1) MKRLHPALPSCLLLVLFGIWRIAPQTHASSAGLPPLSATSFLEDLMDRYGKNDSL
mZipl4b (1) MKRLHPALPSCLLLVLFGIWRTAPQTHASSAGLPPLSATSFLEDLMDRYGKNDSL
Consensus (1) MKRLHPALPSCLLLVLFGIWRTAPQTHASSAGLPPLSATSFLEDLMDRYGKNDSL

(56) 56 70 s0 ,90 100 110
mZipl4a (56) ILTQLKSLLDHLHVGVGRDNVSQPKEGPRNLSTCFSSGDLFAAHNLSERSQIGAS
mZipl4b (56) TLTQLKSLLDHLHVGVGRDNVSQPKECPRNLSTCFSSGDLFAAHNLSERSQIGAS
Consensus (56) TLTQLKSLLDHLHVGVGRDNVSQPKEGPRNLSTCFSSGDLFAAHNLSERSQIGAS

(111) 111 120 130 140 ,150 165
mZipl4a(111) EFQEFCPTILQQLDSQACTSENQKSEENEQTEEGKJPSAIEVWGIGFL -I ,lJ
mZipl4b(111) FQEFCPTILQQLDSQACTSENQKSEENEQTEEGKPSATEVWGIGFL- IE'-
Consensus(111) EFQEFCPTILQQLDSQACTSENQKSEENEQTEEGKPSAIEVWGFGFL VSLI L

(166) 166 180 190 o 0010 220
mZipl4a(166) SLEGV] If i II I- \LIICTLLSNALFQLIPEAFCFNPQDNYVSK
mZip4b (166) SLEGA "-. i ii-- F LlGTLYSNALFQLIPEAFGFNPQDNYVSK
Consensus(166) SLLG VLP K FF RLL YFIALAIGTL SNALFQLIPEAFGFNPQDNYVSK

(221) 221 230 240 250 260 275
mZipl4a (221) SAVVFGGFYLFFFTEKILKMLLKQKNEHHHCHNHFTSETLPSKKDQEEGVTEKLQ
mZipl4b (221) SAVVFGGFYLFFFTEKILKMLLKQKNEHHHGHNHFTSETLPSKKDQEEGVTEKLQ
Consensus (221) SAVVFGGFYLFFFTEKILKNLLKQKNEHHHGHNHF SETLPSKKDQEEGVTEKLQ

(276) 276 290 300 ,310 ,320 330
mZipl4a(276) NGDLDHMIPQHCNSELDGKAPGTDEKVIVNSMSVQDLQASQSACYWLKGVRYSDI
mZipl4b(276) NGDLDHMIPQHCNSELDGKAPGTDEKVIVNSMSVQDLQASQSACYWLKGVRYSDI
Consensus (276) NGDLDHMIPQHCNSELDGKAPGTDEKVIVNSMSVQDLQASQSACYWLKGVRYSDI

(331) 331 340 350 360 P70 385
mZipl4a(331) GTLAWMITLSDGLHNFIDGLAIGA3FTVSVFQGISTSVAILCEEFPHELGDFVIL
mZipl4b (331) GTLAWMITLSDGLHNFIDGLAIGASFTVSVFQGISTSVAILCEEFPHELGDFVIL
Consensus(331) GTLAWMITLSDGLHNFIDGLAIGASFTVSVFQGISTSVAILCEEFPHELGDFVIL

(386) 386 400 410 ,420 ,430 440
mZipl4a (386) LNAGMSIQQALFFNFLSACCCYLGLAFGILAGSHFSANWIFALAGGMFLYIALAD
mZipl4h(386) LNAGMSIQQALFFNFLSACCCYLGLAFGILAGSHFSANWIFALAGGMFLYIALAD
Consensus (386) LNAGMS IQQALFFNFLSACCCYLGLAFGILAGSHFSANWIFALAGGMFLYIALAD

(441) 441 ,450 460 470 489
mZipl4a (441) MFPEMNEVCQEDEKNDSFLVPFVIQNLGLLTGFSIMLVLTMYSGQIQIG
mZipl4b(441) MFPEMNEVCQEDEKNDSFLVPFVIQNLGLLTGFSIMLVLTMYSGQIQIG
Consensus (441) MFPEMNEVCQEDEKNDSFLVPFVIQNLGLLTGF IMLVLTMYSGQIQIG


Figure 4-1. Schematic representation of the mZip14 gene and sequence alignment of its
two protein isoforms. A) The mZip14 gene has 10 exons and encodes two
transcripts due to alternative splicing of equal length of exon 5A or 5B. B) Two
transcripts of mZipl4 gene encode two isoform proteins, named mZipl4a and
mZipl4b. Sequence alignment indicate that two isoforms are 95.9% identical
and 97.5 % conserved (identical amino acids are highlighted in light grey,
dark grey indicates the conservative amino acid exchange).









CHAPTER 4
STRUCTURAL ANALYSIS OF ZIP14

Introduction

ZIP14, a newly identified iron transporter, can mediate both TBI and NTBI into

cells. Although we are starting to understand the transport function of ZIP14, very little is

known about its structure. Determining the structural features of proteins is an essential

part of understanding the basis of their function. ZIPs transport metal ion substrates

across cellular membranes into cytoplasm. Most ZIP proteins are predicted to have

eight transmembrane (TM) regions with a similar membrane topology in which both the

amino- (N-) and carboxy- (C-) terminal ends of the mature peptides are located on the

outside surface of the plasma membrane. ZIP proteins range from 309 to 476 amino

acids in length and this difference is largely due to the length between TM domains III

and IV (178). A cytoplasmic loop between TM domains III and IV is relatively longer and

often contains a histidine-rich domain as a potential metal-binding motif; other loops

between TM regions are quite short (130,178). Features of this eight-TM domain

topology model have been studied for human ZIP1 and ZIP2 (hZIP1 and hZIP2). By

overexpressing hemagglutinin antigen (HA) or 3x HA tagged recombinant hZIP1 or

hZIP2 in K562 cells, it has been shown that both the amino- terminus of hZIP1 and

carboxy- terminus of hZIP2 are extracellular (179,180). Overexpression in HEK 293

cells of mouse Zip4 and Zip5 (mZip4 and mZip5) whose carboxy termini tagged with

HA, indicated that the carboxy termini of both proteins localized extracellularly (181).

However, all these studied ZIPs share no more than 35% consensus positions with

Zip14 (hZIP1, 24.2%; hZIP2, 22.2%; mZip4, 31.5% mZip5, 33.5%). In one study, by

using hydropathy values, minimal charge density, and four programs, hZIP8 which

































2010 Ningning Zhao

PAGE 1

1 FUNCTIONAL AND STRUCTURAL ANALYSIS OF THE METAL -ION TRANSPORTER ZIP14 By NINGNING ZHAO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGRE E OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Ningning Zhao

PAGE 3

3 To my parents, wife and daughter

PAGE 4

4 ACKNOWLEDGMENTS Four years of PhD study is the time I learned most and the writing of this dissertation has been one of the most si gnificant academic challenges I have ever had to face. Without the support and help of the following people, this project would not have been completed. It is to them that I own my deepest gratitude. Dr. Mitchell D. Knutson, who is my major professor. He l ed me into molecular nutrition field and taught me how to be a better scientist He has always been a generous, tireless and constant mentor and supporter His wisdom, knowledge and commitment to the higher standards inspired and motivated me. Dr. Harry S. Sitren, Dr. Bobbi Langkamp-Henken and Dr. Lei Zhou, who are my dissertation committee members. They have provided invaluable support to me over the years. Their advice and guidance for my dissertation work were highly valued and appreciated. Hyeyoung Nam Supak Jenkitkasemwong, ChiaYu Wang, Charlie Michaudet and Stephanie Duarte, who are my colleagues and friends They always help me with various aspects of my experiments and provide valuable advice for my work. I would also like to thank my parents who always support, encourage and believe in me. I sincerely thank my wife, Miaomiao Wu, for your love, support, encouragement and patience.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...................................................................................................... 4 LIST OF TABLES ................................................................................................................ 8 LIST OF FIGURES .............................................................................................................. 9 LIST OF ABBREVIATIONS .............................................................................................. 11 ABSTRACT ........................................................................................................................ 1 4 CHAPTER 1 LITERATURE REVIEW .............................................................................................. 16 Function and Distribution o f Iron in the Body ............................................................. 16 Iron Homeostasis ........................................................................................................ 17 Intestinal Iron Absorption ..................................................................................... 17 Intracellular Storage and Ci rculatory Transfer of Iron ......................................... 21 Iron Utilization by Erythrocytes ............................................................................ 23 Iron Recycling ....................................................................................................... 24 Iron Storage .......................................................................................................... 27 Iron Loss and Excretion ....................................................................................... 27 Role of DMT1 in Iron Metabolism ............................................................................... 27 DMT1 Gene Structure and its Isoforms ............................................................... 28 Role of DMT1 in Iron Absorption ......................................................................... 29 Function of DMT1 in Endosomal Iron Release ................................................... 31 Subcellular Localization of DMT1 ........................................................................ 34 DMT1 is not the Only Transporter Involved in Endosomal Iron Release ........... 36 Iron Uptake by Hepatocytes ....................................................................................... 37 Transferrinbound Iron Uptake by Hepatocytes .................................................. 38 Non -transferrinbound Iron Uptake by Hepatocytes ........................................... 39 ZIP14 and Iron Metabolism ........................................................................................ 40 Identification and Characterization of ZIP14 ....................................................... 40 ZIP14 and Cellular Iron Uptake ........................................................................... 42 2 MATERIALS AND METHODS ................................................................................... 51 Cell Culture ........................................................................................................... 51 Expression of Mouse Zip14 (mZip14) in HEK 293T and HepG2 cells ............... 51 Knockdown of Endogenous ZIP14 in HepG2 Cells Using siRNA ...................... 51 Measurement of TF -bound Iron Uptake .............................................................. 52 pH Dependence of Iron Transport Activity .......................................................... 52 Genetic Knock -in to Tag Endogenous ZIP14 of HepG2 Cells with 3Flag Epitope .............................................................................................................. 53

PAGE 6

6 Western Blot Analysis .......................................................................................... 54 Subcellular Localization ....................................................................................... 55 Quantitative Determination of TF uptake ............................................................ 55 Isolation of Plasma Membrane P roteins by Cell -surface Biotinylation ............... 56 RNA Isolation and Quantitative PCR (qPCR) ..................................................... 56 Construction of GFP Fusion and Flagtagged mZip14 Plasmids ....................... 57 Primary Sequence Analysis and Bioinformatic Predictions ................................ 58 Construction of Asparagine Mutant Plasmids ..................................................... 58 Measurement of Iron Transport Activity .............................................................. 59 Statistical Analysis ............................................................................................... 59 3 ZIP14 AND TRANSFERRIN -BOUND IRON UPTAKE .............................................. 63 Introduction ................................................................................................................. 63 Results ........................................................................................................................ 65 Overexpression of Zip14 Increases TBI Uptake in HEK 293T Cells .................. 65 pH -dependent Iron Transport Activity of Zip14 ................................................... 65 Subcel lular Localization of Zip14-GFP in HepG2 Cells ...................................... 65 ZIP14 and DMT1 mRNA Copy Numbers in HepG2 and HEK 293T Cells ......... 66 Epitope taggin g of Human Endogenous ZIP14 in HepG2 Cells ......................... 66 Subcellular Localization of ZIP14 in HepG2 Cells .............................................. 67 Knockdown of ZIP14 in HepG2 Cells Decreases TBI Uptake ............................ 68 Effect of Holo -TF on the Abundance and Cell -surface Expression of ZIP14 in HepG2 Cells .................................................................................................. 68 Discussion ................................................................................................................... 69 4 STRUCTURAL ANALYSIS OF ZIP14 ........................................................................ 80 Introduction ................................................................................................................. 80 Results ........................................................................................................................ 83 Structure of the murine Zip14 Gene and Sequence Alignment of its Two Protein Isoforms ................................................................................................ 83 Epitope Mapping of mZip14 ................................................................................. 83 Enzymatic Analysis of Glycosylation Status of Zip14 ......................................... 84 Identification of N -linked Glycosylation Sites in mZip14 ..................................... 85 Schematic Membrane Topology Model of mZip14 ............................................. 85 N -linked Glycosylation does not Affect Plasma Membrane Localization of mZip14 .............................................................................................................. 86 N -linked Glycosylation is Required for the Iron Transport Activity of mZip14 .... 86 Discussion ................................................................................................................... 86 5 CONCLUSIONS AND FUTURE DIRECTIONS ......................................................... 99 Conclusions ................................................................................................................ 99 Future Directions ...................................................................................................... 101

PAGE 7

7 LIST OF REFERENCES ................................................................................................. 104 BIOGRAPHICAL SKETCH .............................................................................................. 117

PAGE 8

8 LIST OF TABLES Table page 2 -1 List of Primers Used for PCR to Generate and Screen for Endogenous ZIP14 3 Flag in HepG2 Cells ............................................................................... 61 2 -2 List of Primers Used for Generating 3 Flag-tagged mZip14 Vectors ................. 61 2 -3 List of Primers Used for Generating Asparagine Mutants .................................... 62 4 -1 Bioinformatic Prediction of mZip14 Transmembrane (TM) Regions .................... 98 4 -2 Prediction of mZip14 Signal Peptide Cleavage Site ............................................. 98

PAGE 9

9 LIST OF FIGURES Figure page 1 -1 Distribution of iron within the body ........................................................................ 44 1 -2 Schematic model for intestinal iron absorption .................................................... 45 1 -3 Overview of the ironrespons ive element/ironregulatory protein (IRE/IRP) network. .................................................................................................................. 46 1 -4 Overview of the TF cycle ...................................................................................... 47 1 -5 Overview of the receptor mediated pathway for endocytosis of extracellular heme and hemoglobin ............................................................................................ 48 1 -6 Schematic demonstration of mutations in the iron transport protein DMT1 ......... 49 1 -7 Iron uptake by hepatocytes. ................................................................................... 50 3 -1 Overexpression of Zip14 increases TBI uptake. ................................................... 73 3 -2 pH dependence of Zip14 mediated iron transport. ............................................... 74 3 -3 Subcellular localization of Zip14-GFP in HepG2 cells. ......................................... 75 3 -4 Comparison of ZIP14 and DMT1 mRN A levels in HepG2 and HEK 293T cells. ........................................................................................................................ 75 3 -5 Targeted knock -in of 3 Flag into the ZIP14 locus of HepG2 cells ...................... 76 3 -6 Confocal mi croscopic analysis of the subcellular localization of ZIP14 in HepG2 cells. ........................................................................................................... 77 3 -7 Knockdown of ZIP14 decreases TBI uptake by HepG2 cells ............................... 78 3 -8 Effect of holo -TF on the abundance and cell -surface expression of ZIP14 in HepG2 cells. ........................................................................................................... 79 4 -1 Schematic representation of the mZip14 gene and sequence alignment of its two pr otein isoforms. .............................................................................................. 90 4 -2 Flag epitope mapping of mZip14. .......................................................................... 91 4 -3 The carboxy terminus of mZip14 is intracellular. .................................................. 92 4 -4 Deglycosylation of Zip14. ....................................................................................... 93 4 -5 Identification of N -linked glycosylation sites in mZip14. ........................................ 94

PAGE 10

10 4 -6 Schematic illustration of mZip14 membrane topology. ......................................... 95 4 -7 N -linked glycosylation does not affect plasma membrane localization of mZip14. ................................................................................................................... 96 4 -8 Functional analysis of mZip14 lacking N linked glycosylation. ............................. 97

PAGE 11

11 LIST OF ABBREVIATION S AAV Adeno associated virus BSA B ovine serum albumin bp Base pair Cp Ceruloplasmin Dapi 4',6-diamidino -2 phenylindole Dcytb Duodenal cytochrome B DMEM Dulbeccos modified Eagles medium DMT1 D ivalent metal transporter 1 EPO Erythropoietin ER Endoplasmic reticulum FPN1 Ferroportin 1 GFP Green fluorescence protein GI G astrointestinal Hb Hemoglobin HCP 1 Heme carrier protein 1 HEPES 4 -(2 hydroxyethyl) -1 piperazineethanesulfonic acid HFE Gene mutated in the most common form of hemochromatosis HH Hereditary hemochromatosis Hp Haptoglobin HO H eme oxygenase HRP H orseradish peroxidase Hx H emopexin Irt Iron -r egulated t ransporter IRP Iron -regulatory protein

PAGE 12

12 IR E Iron -responsive element k Da Kilodalton LAMP1 Lysosomal associated membrane protein 1 LAMP2 Lysosomal associated membrane protein 2 LDL L ow density lipoprotein LPS L ipopolysaccharide LRP LDL receptor -related protein mRNA Messenger RNA Fe -NTA Ferric nitrilotriacetic acid NTBI Non -transferrinbound iron PFA P araformaldehyde RBC Red blood cell RES R eticulo endothelial system ROS R eactive oxygen species SFM Serum -free medium siRNA Small interfering RNA Slc /SLC Solute carrier Steap Six transmembrane epithelial antigen of the prostate TBI Transferrinbound iron TBS Tris -buffered saline TBST Tris -buffered saline Tween -20 Tf/ TF Transferrin TfR1/ TF R1 Transferrin receptor 1 TfR2/ TF R2 Transferrin receptor 2 TM Transmembrane

PAGE 13

13 TRP T ransient receptor potential cation chan nel ORF Open reading frame PBS Phosphate buffered saline PNGase F P eptide N glycosidase F qPCR Quantitative polymerase chain reaction UTR Untranslated region WT Wild -type Zip/ZIP Z rt and I rt -like p roteins Zrt Z inc -r egulated t ransporter

PAGE 14

14 Abstract of Diss ertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy FUNCTIONAL AND STRUCTURAL ANALYSIS OF THE METAL -ION TRANSPORTER ZIP14 By Ningning Zha o August 2010 Chair: Mitchell D. Knutson Major: Nutritional Science s ZIP14 ( solute carrier family 39, member 14, SLC39A14) is a transmembrane metal -ion transporter that is abundantly expressed in the liver, heart, and pancreas. Previous studies of HEK 2 93 cells and the hepatoc yte cell lines AML12 and HepG2 established that ZIP14 mediates the uptake of non-transferrinbound iron (NTBI), a form of iron that appears in the plasma during iron overload disorders I investigated the role of ZIP14 in transferri n -bound iron (TBI) uptake and determined the subcellular localization of ZIP14 in HepG2 cells. It was found that overexpression of mouse Zip14 (mZip14) in HEK 293T cells increased the uptake of TBI without increasing levels of transferrin receptor 1 (TFR1) I also found that mZip14 can mediate the transport of iron at pH 6.5, the pH at which iron atoms dissociate from transferrin (TF) within endosomes I used an adenoassociated viral (AAV) system to generate a HepG2 cell line expressing a Flag-tagged ZIP14 allele allow ing specific and sensitive detection of endogenous ZIP14 in these cells Confocal microscopic analysis detected ZIP14 at the plasma membrane and in TF -containing endosomes Knockdown of endogenous ZIP14 with siRNA did not decrease TFR1 or TFR 2 levels, but resulted in a 45 % reduction in

PAGE 15

15 TBI uptake without decreasing TF uptake in HepG2 cells These results suggest that ZIP14 participates in the uptake of iron from TF thus identifying a potentially new role for ZIP14 in iron metabolism. Most ZIP proteins have been predicted to contain eight transmembrane (TM) helices with both amino (N -) and carboxy (C -) termini located extracellularly. Structural analysis by Flag epitope mapping and bioinformatic prediction s upport a model of mZip14 that has s even TM segments It was also found that t he N terminus is an ectodomain, wh ereas the C -terminus and the long loop containing a histidine -rich putative metal -binding motif localize intracellularly. In addition, N -linked glycosylation sites were identifie d through site directed mutagenesis, and it was found that mZip14 is glycosylated at asparagines 52, 75, 85 and 100, residues that are all in the extracellular N terminus. It was also demonstrated that N -linked glycosylation is not required for cell surfa ce localization but it is requir ed for iron transport activity. In conclusion, a new role of ZIP14 in iron metabolism was identifiedZIP14 m ediates the uptake of iron bound to TF I also investigated the membrane topology of m Zip14 and conclude that it has seven transmembrane regions. Finally, I found that N linked glycosylation is essential for the iron transport ability of Zip14 but not for its trafficking to the plasma membrane.

PAGE 16

16 CHAPTER 1 LITERATURE REVIEW The first part of this chapter will provide a general overview of the function and distribution of iron in the body and the regulation of iron homeostasis. The second part will focus on the role of the transmembrane iron transport protein, divalent metal transporter 1 ( DMT 1 ) in cellular iron uptake. The last section will introduce ZIP14 and its relationship with cellular iron metabolism. F unction and D istribution o f Iron in the B ody Iron is the second most abundant metal and the fourth most abundant element in the earths crust (1) The study of iron biology focuses on how cells and organisms regulate their iron content, how various tissues coordinate iron distribution, and how dysregulated iron homeostasis leads to common metabolic hematological and neurodegenera tive diseases (2) As an essential nutrient for nearly all living cells, iron is utilized in many biological functions. By being a transition metal, iron has the ability to accept and donate electrons easily, interconverting between ferrous (Fe2+) and ferric (Fe3+) states. This charact er makes iron valuable as a component of oxygen -binding molecules (e.g. hemoglobin and myoglobin), cytochromes in the electron transport chain and an element in a variety of enzymes, especially those containing heme. The same redox potential accounts for i ts damaging effect s through catalyzing the Fenton reaction, in which H2O2 is converted to the highly reactive hydroxyl radical ( OH ). Hydroxyl radicals oxidatively modify fatty acids, proteins, and nucleic acids leading to cellular dysfunction (3) The distribution o f body iron in normal adult s is illustrated in Figure 11 Male adults typically have 35 to 45 mg of iron per kilogram of body weight. Premenopausal

PAGE 17

17 female s have less iron store s due to periodic blood loss through menstruation (4) In a balanced state, approximately 0.5 to 2 mg of dietary iron is absorbed through duodenal ent erocytes every day, and the same amount is lost in the urine, feces, sweat and sloughed cells (5,6) About 65% of the iron in the body is incorporated into hemoglobin of red blood cells (RBCs) and 10% is present in myoglobin, other enzymes and cytochromes. The remaining body iron is stored in the liver, macrophages of the reticuloendothelial system (RES) and bone marrow (4,7,8) Iron H omeostasis Regulation of systemic iron hom eostasis involves intricate control of intestinal iron absorption, effective erythropoietic iron utilization, efficient iron recycling from effete erythrocytes, and controlled iron storage by hepatocytes and macrophages (9) Intestinal Iron A bsorption Since deficiency of iron results in anemia, while overload leads to formation of reactive oxygen species (ROS), ir on homeostasis needs to be tightly controlled. The systemic regulation of iron homeostasis take s place primarily in intestinal iron absorption because humans have no physiologic pathway to excrete excess iron (4). Dietary iron is found in tw o basic forms, either as heme, found in meat and meat products or non-heme iron, prese nt in vegetables, beans, and fruits. Nonheme iron predominates in all diets, comprising 90% 95% of total daily iron intake. The regulation of iron absorption relies on mechanisms that sense dietary iron content as well as iron storage levels in the body and erythropoietic iron requirements (10) The absorption of iron is illustrated in Figure 1 -2. Dietary iron uptake occurs at the apical membrane of duodenal enterocytes. The i nsoluble ferric form of iron is the primary non-heme iron in food and must be reduced to fer r ous iron before transporting across the intestinal

PAGE 18

18 epithelium. Cytochrome b-like ferrireductase (Dcytb) is thought to be the responsible enzyme (11) However, by Dcytb gene knockout in mice, Gunshin et al. (12) demonstrated that the absence of Dcytb did not impair accumulation of body iron stores when mice were fed normal chow diet, indicating that there was no maj or effect on intestinal absorption I t was concluded that Dcytb is dispensable for intestinal iron absorption in mice. However the role of Dcytb in intestinal iron absorption remains unclear as direct iron absorption was not measured in th e knockout study (13) Moreover mice are capable of synthesizing ascorbic acid and may have less need for a duodenal surface ferric reductase (10) The r ecently identified six -transmembrane epithelial antigen of the prost ate 2 (Steap2) protein might be another candidate for this ferrireductase role (14) There are also various dietary components, such as ascorbic acid, cysteine and histidine which are capable of reducing Fe3+ to Fe2+ (15 -17) After being reduced by Dcytb or other reducing agents, Fe2+ becomes a substrate for a transmembrane transporter. Gunshin et al (18) identified in rats a metal -ion transporter named DCT1 (divalent cation transporter 1, also called DMT1, divalent metal transporter 1) and found that this protein which is upregul ated by dietary iron deficiency, may represent a key mediator of intestinal iron absorption. In Caco 2 cells (human epithelial colorectal adenocarcinoma cells) it has been shown that DMT1 is expressed in the apical membrane (19) Non vegetarian diets contain another fo rm of iron, heme iron (primarily from hemoglobin and myoglobin), which may account for 10% of the dietary iron intake. Because heme iron is more bioavailable than nonheme iron, it may contribute half of the total iron absorbed in a western meat -rich diet (20) Absorption of heme occurs by

PAGE 19

19 binding to the duodenal brush border membrane (20) Heme carrier protein 1 (HCP1) was identifie d to be a heme import protein (21) but subsequent studies showed that it was a folate transporter (22,23) T he precise role of HCP1 in iron metabolism will require further i nvestigation Following uptake into enterocytes, heme is broken down by heme oxygenase 1 (HO); l i berated ferrous iron enters the intracellular iron pool with the iron absorbed from non-heme pathway (24) Absorbed iron has two fates depending on the bodys requirements. Some of the iron extracted from the diet is stored in ferritin within the enterocytes, wh ereas some is exported across the basolateral membrane of enterocytes. If body iron stores are sufficient, and there is no increased erythropoietic demand, a large amount of newly absorbed iron will be stored as ferritin in enterocytes. This intracellular iron is lost with in 3 to 4 days as epithelial cells slough into the intestinal lumen (25,26) The t ransfer of iron across the basolateral membrane of enterocytes into the circulation is accomplished by the coordination of a transport protein ferroportin1 (FPN1) and a ferrioxidase, hephaestin (27,28) FPN1 is the only iron exporter identified to date. Ferrous iron is exported across the basal membrane by FPN1 and then oxidized by hephaestin before becoming bound to plasma TF. I ntestinal iron absorption is regulated by body iron stores, recent dietary iron intake and erythropoietic activity. Two models have been proposed to explain how the a bsorption of iron is regulated: the crypt programming model and the hepcidin model (7,8) The first one is based on a post -transcriptional regulation scheme of iron metabolism in which c ellular iron levels are balanced mainly by the iron-responsive element/iron regulatory protein (IRE/IRP) system (29)

PAGE 20

20 Role of the IRE/IRP regulatory system in cellular iron homeostasis: Cellular iron homeostasis is attained by coordin ated and balanced expression of proteins involved in iron uptake, export, storage and usage. The posttranscriptional regulation mediated by the IRE/IRP system appears to be essential, which is a specific mRNA protein interaction in the cytoplasm (2) An overview of IRE/IRP regulatory network is illustrated in F igure13. In this system, particular cis -regulatory elements, stem loop structures cal led iron -responsive elements (IREs) in the untranslated regions (UTR) of respective mRNAs, are recognized by trans acting proteins, known as iron -regulatory proteins (IRPs), which are cytosolic iron sensors that control the rate of mRNA translation or stability (30) Two IRPs (IRP1 and IRP2) have been identified. Both IRPs belong to the aconitase family proteins and share 64% amino acid sequence identity, but different mechanisms control their activities in response to cellular iron levels (31) During irondeficient conditions, IRP1 binds IREs. Binding at the 5 UTR blocks ribosome access t o mRNA, inhibiting translation, whereas binding at the 3 UTR leads to increased mRNA stability by preventing mRNA degradation. Under iron-sufficient conditions, an iron -sulfur cluster (4Fe 4S) assembles in IRP1, preventing its binding to IREs. During iron-deficiency, IRP1 is open and accessible for interaction with IREs. In co ntrast, IRP2 does not contain the iron -sulfur cluster and regulation by iron is through proteasomal degradation (32,33) The crypt programming model: T his model proposes that the intracellular iron level of the crypt cells correspond s to the bodys iron stor e level which in turn regulate s dietary iron absor ption via mature villus enterocytes (7,8) Iron uptake from plasma through TBI into crypt cells determines the iron content within the cells The intracellular

PAGE 21

21 iron concentration controls the interaction of cytosolic IRPs with IREs in the 5 and 3 UTRs of different mRNA molecules. Under low iron conditions, IRP1 binds to IREs of DMT1 mRNA at 3 UTR ends, the transcript is stabilized, translation proceeds, and the pr otein is synthesized. Thus, low body iron stores results in upregulation of DMT1 in the duodenum and increased dietary iron absorption. The hepcidin model: The l iver secret e s hepcidin, a 25amino acid cysteine-rich peptide with antimicrobial properties (34) Hepcidin expression is regulated by a number of factors such as inflammation, hypoxia, anemia and liver iron stores. Hepcidin has been identified as an important regulator of iron homeostasis (35) The hepcidin mod el proposes that hepcidin is secreted into th e blood and interacts with vill us enterocytes to regulate dietary iron absorption. This regulation is through controlling the expression of FPN1 at the basolateral membrane. The binding of hepcidin to cell -surfa ce FPN1 causes internalization and degradation of FPN1. During iron deficiency, anemia or hypoxia, hepcidin expression is reduced, causing an increase in FPN1 expression. As a consequence, d ietary iron release in to plasma from the basolateral side of intes tinal cells increases Most evidence supports the hepcidin model as the more likely one. Intracellular S torage and C irculatory Transfer of I r on To avoid the harmful effects of free iron during oxidative stress, iron is usually bound to specialized protei ns, such as intracellular ferritin and circulating TF Ferri tin is present in every cell type. Sm all quantities of ferritin, derived from macrophages (36) are also present in human se rum and are elevated in conditions of iron overload or inflammation. S erum ferritin is widely used as a clinical indicator of body iron stores (37)

PAGE 22

22 Ferritin is a 450 kDa, cagelike heteropolymer of 24 subunits, including H (heavy, 21 kDa) chains and L(light, 19 kDa) chains (38) This structure makes it capable of storing up to 4500 Fe3+ atoms, sequestering iron in a nontoxic form (39) The genes encoding for H and Lferritin reside on different chromosomes and are expressed independently. The H subunit is capable of oxidizing Fe2+ to Fe3+, whereas the L subunit assist s in hydrolysis and core formation. Ferr itin functions as a buffer against iron deficiency and overload, as it stores and releases iron in a controlled manner. When iron levels are low, ferritin synthesis is decreased, less iron is stored and more can be utilized by the cells; conversely, when i ron levels are high, ferritin synthesis increases, protecting cells from damage by excess iron (37) This regulation is mainly through the well -characterized post -transcriptional, irondependent machinery based on the interaction between IRPs and IREs The IRE within ferritin is a stem loop st ructure localized in the 5 UTR of both H and Lferritin mRNA. When iron is abundant, IRP 1 assembles 4Fe4S cluster and functions as cytosolic aconitase; during iron deficiency, the cluster is disassembled and the protein binds IRE with high affinity (40) In the case of ferritin, when IRP binds to its 5 UTR, translation is repressed (Figure 1 3). T F is a glycoprotein produced mainly in the liver with a molecular mass around 80 kDa. It contai ns two specific high affinity ferric iron binding sites located in its aminoand carboxy terminus (1) Th e binding of TF to iron atoms is associated with binding of an anion, usually bicarbonate (41) It has been established that TBI can be released by acidification (42) Mammalian cells take up TBI through receptor mediated endoc ytosis. TFR1 is a homodimer of a 90 kDa glycosylated polypeptide with 61 amino acids N terminal of each subunit local ized in the cytoplasm (4345) This intracellular tail directs

PAGE 23

23 rapid internalization of T F -T F R 1 complex upon binding of the holo -TF ( diferric TF ) to its receptor with formation of a coated pit and a coated vesicle which sheds its clathrin to become an endosome. Once internalized, endosomes are acidified by an ATP dependent proton pump (46) Acidification weak ens binding of ferric iron to TF and causes c onformational changes of both TF and T F R 1 (47) Dis sociated ferric iron is reduced to the ferrous form by Steap3 (six -transmembrane epit helial antigen of the prostate 3) the dominant f errireductase in the erythroid TF cycle (14,48) After reduction, Fe2+ is transported into the cytosol by DMT1. Both TF and TFR1 will return to the cell surface, where TF is released for further usage. The TF cycle is depicted in Fig ure 1 4 Iron U tilization by Erythrocytes Erythrocytes require iron for the oxygen -carrying capacity of hemoglobin. The erythroid bone marrow is the largest consumer of iron. Failure to incorporate adequate iron into heme results in i mpaired erythrocyte maturation and lead s t o microcytic hypochromic anemia. In healthy individuals, about two-thirds of the to tal body iron is accounted for by hemoglobin in developing erythroid precursors and mature RBC s. Approximately 20 25 mg of iron are needed every day for hemoglobin synthesis (26) Nearly all circulating iron is bound to TF in the plasma. T F carries one or two atoms of iron per protein molecule. Erythroid precursors meet their iron needs by taking up TBI through the TF cycle (26) The amount of TF R 1 present on the cell surface determines the amount of iron imported into cells. The expression of TF R 1 is regulated developmentally during erythoid maturation, correlating with the changing rates of hemoglobin production (49)

PAGE 24

24 Murine models reveal the important role of Tf / TfR 1 -mediated iron uptak e for erythropoiesis. Firstly, h ypotransferrinemic ( Trfhpx/hpx) mice carry a spontaneous mutation in the T f gene (50) These mice are T f deficient, with only a bout 1% of normal circulating Tf concentrations They develop severe microcytic hypochromi c anemia, ind icating an indispensable role of Tf in iron delivery to developing erythroid cells (51) Knockout mice have also been produced which lack functional Tf R1 protein (52) These mice die in utero bet ween embryonic day 9.5 and 11.5, apparently as a consequence of severe anemia. The heterozygous mice lacking one copy of the TfR1 gene demonstrate iron-deficient erythropoiesis, even though they retain one normal TfR1 allele. Iron R ecycling The major source of plasma iron does not come from intestinal absorption, but from macrophages that recyc le iron from senescent or damaged erythrocytes (53) Macrophage iron recycling is quantitatively important because the amount of iron supplied by this way each day is about 20 times greater than the amount absorbed through the small intestine (2) Iron recycl ing takes place in the RES, a term that describes specialized macrophages that are found in the liver (Kupffer cells) the spleen and the bone marrow (53) Binding of erythrocytes to the macrophage cell surface initiates phagocytosis and lysosomemediated degradation of effete RBCs Heme is liberated and catabolized by H O releasing inorganic iron biliverdin and carbon monoxide (54) Free iron is either stored in ferritin or released into the circulation through the iron export protein FPN1 FPN1 is highly expressed in extraembryonic visceral endoderm cells which are important for providing nourishment to the developing embryo. Global inactivation of the murine FPN 1 gene leads to embryonic lethality due to a defect of the developi ng embryo to acquire iron, indicating that FPN 1 is essential in

PAGE 25

25 early development (55) FPN1 is also highly expressed in the intestinal basolateral epithelium and in macrophage s of the liver and spleen. FPN1 levels increase markedly in macrophages after erythrophagocytosis (56) When FPN 1 is inactivated in the embryo proper, sparing the extraembryonic visceral endoderm, mice survive, but become anemic shortly after birth when they must rely on intestinal absorption f or iron accumulation (55) Also, macrophages of the liver and spleen showed marked iron retention under this circumstance, likely because of an inability to release recovered i ron from phagocytosed effete erythrocytes. These observations suggest that FPN1 is the major iron exporter functioning in iron-recycling macrophages. Normal adult plasma contains 3-4 mg of iron, almost all bound to TF Most of circulating iron is from RES recycling of RBC iron; only small amounts are contributed by hepatic iron stores and intestinal absorption. But, e rythrophagocytosis is not the only way for macrophages to acquire iron. They can obt ain iron from TF ( 57) O ther pathways in macrophages t ake up cytotoxic free heme and hemoglobin (Fig 1 5 ). Hemoglobin is the most abundant and functionally important protein in erythrocytes. But once released from RBC s, it becomes highly toxic because of the oxidative properties of heme (protoporphyrin IX and iron) via the Fenton reaction to produce ROS causing cell injury or death (58) Normally, only a small amount of extravascular hemolysis occur s due to destruction of senescent erythrocytes, causing Hb release into plasma. Yet under various intravascular hemolysis linked conditions, such as hemorrhage, hemoglobinopathies, ischemia reperfusion, sickle cell disease or malaria, large amounts of free Hb are released (58) Hemoglobin released into plasma is readily bound

PAGE 26

26 stoichiometrically by the liver derived plasma protein haptoglobin (Hp), forming a h emoglobinhaptoglobin (Hb-Hp) complex (59) CD163, a hemoglobin scavenger receptor present on the surface of monocytes and macrophages in the liver and several other tissues mediate s the endocytosis and subsequent degradation of Hb-Hp complex (60) T he entire Hb -Hp complex is degraded in lysosomes to release heme and various proteolytic products within the cells. The heme is then degraded by heme H O 1 to release iron, carbon monoxide, and biliverdin. The free iron enter s the same intracellular pool as iron taken up from other sources. Mice in which the haptoglobin gene is inactivated do not have obvious disorder s in iron metabolism, suggesting that Hp does not play a major role in normal iron metabolism (61) Plasma Hb is quickly oxidized to ferrihemoglobin when the binding capacity of Hp is exceeded (58) Ferrihemoglobin dissociates into globin and ferriheme. Ferriheme is transferred to hemopexin (Hx), forming Heme-Hx complex. This complex is internalized by receptor mediated endocytosis into macrophages, which express LDL receptor related protein (LRP)/CD91, the receptor for the heme-Hx complex (62) LRP -CD91mediated endocytosis is also a degradation pathway for all known ligands of this receptor (63) In addition to macrophages, LRP/CD91 is highly expressed in several other cell types including hepatocytes and neurons (62) Inactivation of the Hx gene in mice does not lead to disturbed iron metabolism, suggesting that the role of Hx in iron homeostasis is important only under pathological conditions (64) Hemopexin levels in serum reflect how much heme is present in the blood. Therefore, a high Hx level indicates that there has been significant degradation of heme-containing compounds and low Hx levels are one of the diagnostic features of hemolytic anemia. Extracellular

PAGE 27

27 heme can be cleared through these two receptor mediated endocytosis pathways, thus, avoiding the strong oxidative features and proinflammatory effects of free heme. Iron Storage The liver is t he main storage organ for iron and accounts for approximately 5 0% of the storage iron (65,66) In rats about 98% of iron stored in the liver is found in hepatic parenchymal cells and most of the remainder is found in Kupffer cells, with only very small amounts in stellate cells, endothelial cells and bile duct cells (67) Iron is primarily stored in macrophages of the RES and in hepatocytes of the liver Iron released from metabolized heme can be stored in two forms within the cell: as ferritin in the cytosol and as hemosiderin after ferritin breakdown in lysosomes (8) Hemosiderin in macrophages inc reases dramatically in iron overload, but only represents a small portion of normal body iron stores. Tissue macrophages, particularly in the spleen and liver, also express ferritin, which can store iron that is not needed elsewhere. Iron L oss and E xcretio n Although i ron homeostasis is tightly controlled in terms of uptake, recycling and utilization, iron loss and excretion are not actively regulated. Iron is lost from the body by sloughing of mucosal cells, desquamation of skin cells, blood loss, sweat and urinary excretion. The amount of iron lost every day is about 1 to 2 mg, equivalent to the amount taken up through normal daily diet (6,68) Role of DMT1 in Iron M etabolism Divalent metal transporter 1 (DMT1 or S LC39A2 solute carrier family 11, member 2 ) is the most well -characterized transmembrane iron transport protein.

PAGE 28

28 DMT1 Gene Structure a nd its Isoforms DMT1 was first identified in 1995 when screening for murine homologs of Nramp1 (natural resistance associat ed macrophage protein 1) a protein involved in host defense. Accordingly, DMT1 was first named Nramp2. Nramp1 is an integral membrane phosphoglycoprotein expressed in lysosomes of macrophages. It is targeted to the membrane of the phagosome after phagocyt osis (69) Howe ver, the ubiquitous expression of DMT1 mRNA did not seem particularly associated with organs or cells implicated in host defenses and it localized to chromosome 15, which is away from known host defense genes (70) The DMT1 gene encodes f our different protein isoforms. The upstream 5 exon, termed exon 1A, adds an in-frame translation initiation codon and extends the open reading frame (ORF) of the protein by 29 to 31 amino acids in different species (71) The sequences of the variants with or without exon1A are identical after reaching exon 2, and until reaching the C terminal sequences of the ORF P rocessing of the 3 end also yields two DMT1 mRNA variants that differ in their ORF sequences and th eir adjacent 3 U TR s One contains the stem -loop structure, IRE ( iron response element ), the other does not. The n on -IRE isoform has a 25amino acid segment replacement at the C -termin us for 18 amino acids of the IRE -containing isoform (72) F our isoforms are produced by alternative splicing of the 5 end exon 1A and the 3 end IRE, yielding DMT -1A(+ IRE ), DMT -1B(+ IRE ), DMT -1A (-IRE ) and DMT -1B (-IRE ) isoforms (71) Similar to t he IREs in the 3' UTR of the TFR1 mRNA, IRP would bind under low iron conditions and stabilize the DMT1 mRNA, leading to an increased level of DMT1 protein (73) However, in Caco2 cells, it has been shown that the 3 end IRE is not mandatory for iron regulation and the presence of exon 1A is itself associated with iron regulation

PAGE 29

29 (71) The t issue distribution profile indicates that the exon 1B isoform is ubiquitously expressed, whereas t he expression of exon 1A iso form is tissue -specific and particularly abundant in the duodenum and the kidney The same s tudy also found that, in mice, DMT1 regulation in the kidney is associated with the presence of an IRE in the 3 UTR, whereas in the duodenum, iron regulation is most strongl y associat ed with the presence of exon 1A. The multiple isoforms of DMT1 clearly add to the complexity for studying the expression, function, and regulation of this protein. Role of DMT1 in Iron A bsorption In rodent models, such as Belgrade (b ) rats and mk mice mutation of a single nucleotide in the DMT1 gene results in a substitution of a rg inine for gly cine at position 185 (G185R) leading to anemia and iron deficiency (74,75) The iron deficiency results from impa ired intestinal iron absorption and erythroid iron utilization. The p henotype of b rat s was first described in 1966, when i nherited anemia was rep orted among the offspring of X ray -irradiated rats in a nuclear science laboratory in Belgrade in ex Yugoslavi a (76) The Belgrade (b ) laboratory rat suffers from anemi a accompanied by elevated plasma iron and ironbinding capacity, decreased stainable iron in tissues and decreased growth rate. Oats et al. (77) showed that the reduced uptake of both ferric and ferrous iron in homozygous b rats probably involves a defective iron carrier associat ed with the apical m embrane of the duodenum. Using a positional cloning approach, Fleming et al. (74,75) identified DMT1 as the defective gene for both b rats and mk mice Injection of RNA synthesized from DMT1 cDNA in Xenopus oocytes promoted the uptake of iron as well as other divalent cations, including manganese, cobalt, and zinc (18) Uptake of metals by DMT1 is pH dependent, involving proton symport I n Caco2 cells, DMT1 was shown to transport iron

PAGE 30

30 preferentially over other divalent cations (19) Studies in other cultured mammalian cells have also demonstrated that DMT1 can transport a variety of divalent cations at the plasma membrane, including iron (74,78) In addition, it has been shown that iron uptake was most efficient in COS 7 ( a cell line derived from kidney cells of the African green monkey ) and HEK 293 (human embryonic kidney cells) cells transfected with the DMT1 (-IRE) isoform at pH 5.5 6.5, whereas iron uptake decreased to almost baseline level at pH 7.5 or above (79) Overexpression of DMT1 b (G185R) and wild-type construct s in HEK 293T cells confirmed the loss of iron transport function of the G185R mutant, supporting a role for DMT1 in iron metabolism and str engthening the hypothesis that this mutation accounts for the b phenotype (80) In fact, when the DMT1 G185R mutant was stably expressed in cell lines, multiple aspects were altered, including retention in endoplasm ic reticulum (ER), abnormal glycosylation, rapid degradation by a proteasome-dependent mechanism, and less active metal transport activity (81) Furthermore, mk mice show ed a significant increase in the e xpression of G185R mutant in the duodenum (both mRNA and protein levels) but little of the protein was detected at the brush border of enterocytes, suggesting that the G185R mutation impairs not only transport, but also membrane targeting of the protein i n mice (82) In rats fed irondeficient diet, DMT1 m RNA was s ignificantly induced in duodenal epithelial cells as detected by Northern blotting (18) In situ hybridization with a label ed DMT1 cRNA probe in small intestine indicated that DMT1 was highly expressed in enterocyte villus Such localization is consistent with the major site for iron absorption. Dietary iron deficiency results in a dr amatic upregulation of the DMT1(+ IRE ) form in

PAGE 31

31 proximal duodenum (18,83) Localization of DMT1 is in agreement with the known physiological site for ferrous iron absorption in the intestine, which is mostly restricted to the brush border of the proximal intestine (84) An imm unohistochemical study of h uman duodenum showed that DMT1 localize s to enterocytes especially at the microvillus brush border membrane (85) Also, preincubation with DMT1 antibodies significantly inhibited iron upt ake in Caco2 cells at pH 5.5 (19) Mice with specific inactivation of DMT1 gene in the intestine were born alive, but they rapidly developed iron-deficiency anemia (86) The phenotype of the intestine-specific DMT1 knockout mouse firmly establishes that DMT1 is the major iron transporter in the small intestine. Function of DMT1 in Endosomal I ron R elease In Belgrade ( b ) rats, iron uptake from TF by erythropoietic cells is dimin ished and globin synthesis is defective (76,87) A s mall decrease in endocytosis of TF associated with diminished iron uptake and increased iron release by exocytosis from b rats indicated that the defect lies in the metabolism o f TF -iron after its endocytosis (88) It has also been shown that diferric -T F is taken up into b reticulocytes, but iron is poorl y retained, and much is recycled to the extracellular space along with T F meaning that b reticulocytes are unable to move iron out of the vesicle after endocytosis (89) Furthermore, an apparent deficit was also observed with NTBI uptake into b rat erythroid cells, indicating that the b defect is not simply due to a failure to dissociate iron from T F (90) Given the evidence that endosomal iron transport is not efficient in the b rat and both b rats and mk mice have the identical G185R mutation in DMT1 the hypothesi s was formed that DMT1 is the TF cycle endosomal iron transporter in additi on to an intestinal iron transporter (74)

PAGE 32

32 Several studies have reported cases of DMT1 mutations in humans Mims et al. (91) first report ed a case of a female with severe hypochromic microcytic anemia and ir on overload. They found a G -to -C mutation in exon 12 (DMT1 1285 G C), r esulting in a conservative glutamic acid to aspartic acid ( E399D ) single amino acid substitution H owever, the predominant effect of this mutation is exon 12 skipping during processing of the mRNA present in erythroid cells. Removal of exon 12 delet es transmembrane domain 8, which may interfere with proper protein insertion into the membrane. It has been shown that E399D resides in the 4th predicted intracellular loop of DMT1 and forms part of a highly conserved transport signature motif among specie s (92) By expressing mutant E399D, E399Q, E399A in LLC -PK1 kidney cells ( epithelial cell line derived from porcine kidneys ), it has been found that the mutants are fully functional in term of stabi lity and targeting to the membrane, and are also transport -competent, indicating that DMT1G1285C is not a complete loss of function. In fact, G 1285C mutation greatly increase s the exon 12 skipping risk (90% compared with 10% in healthy individuals and 50% heterozygote phenotypically normal sibling). The l imited amount of functional E399D produced in the DMT1G1285C patient may be adequate for iron absorption, but not sufficient for efficient utilization in the erythrocytes. Thus the clinical phenotype appears to be primarily due to exon 12 skipping rather than one amino acid substitution. Interestingly liver biopsy indicated that this patient had severe iron loading in both hepatocytes and Kupffer cells, suggest ing the existence of another iron transport pa thway functioning in the liver Iolascon et al. (93) repor ted an infant with hypochromic, microcytic anemia, and hepatic iron overload with increased serum iron, TF saturation, and serum ferritin levels

PAGE 33

33 before treatment with recombinant erythropoietin (rEpo ). No mutations were found in the HFE gene ( the gene mutated in hemochromatosis) suggesting a defect in iron usage. Screening for DMT1 identified compound heterozygosity for mutations including two novel mutations: a 3bp deletion in intron 4, resulting in a sp l icing abnormality and a C T transition in exon 13 causing the substitution of arginine with cysteine (R416C) in DMT1. At age 5, this patient display ed severe iron loading in the liver despite l ow serum ferritin level s The high iron store s were disproportionate to the iron from transfusions or bone marr ow redistribution, indicating a significant increase of intestinal absorption. Administration of r Epo resulted in rapid improvement of anemia, suggesting the effect of Epo on erythropoi esis and the existence of a DMT1-independent pathway for iron utilization by the TF cycle Beaumont et al (94) identified compound heterozygosity for an in-frame deletion and a substitution mutation in the DMT1 gene in a 6year old French girl representing the third case of congenital microcytic hypochromic anemia due to DMT1 mutations. This patient ha d two mutations in DMT1 including a deletion of a GTG codon in exon 5 and a glycine to valine (G212V ) mutation in exon 8, resulting in the in -frame deletion of V114 (delV114) in TM2 and a G212V substitution in TM5 respectively. Different from two previous reported DMT1 mutations, the anemia was less severe in this case, indicating partial lo ss of function. Figure 16 illustrates the identified DMT1 mutations in rodents and human. In summary, b ased on the observation that hypotransferrinemic ( Trfhpx/hpx) mice and congenital atransferrinemia ( Trf/ -) patients have a severe microcytic hypochromic anemia and high serum iron, it can be concluded that erythroid pre cursors in the bone

PAGE 34

3 4 marrow rely primarily on the endosomal TF -TFR1 pathway for iron uptake. Rapid develop ment of liver iron overload in these animals indicates that insufficient ut i lization of iron in erythrocytes tri g gers the increased absorption of iron from intestine. The i dentical mutation in DMT1 of b rats and mk mice, together with three reported cases of DMT1 mutations in human s, emphasize s the role of DMT1 in erythropoiesis (94) I n addition, DMT1 is express ed in bone marrow and optimal ly functions at acidic pH such as in endosomes (18) These observations support the rational e that DMT1 is an endosomal iron transport protein, especially in erythroid cells. Subcellular L ocalization of DMT1 Studies of the subcellular localization of DMT1 have been done mainly by categorizing DMT1 into the +IRE isoform and the -IRE isoform DMT1( + IRE) protein is expressed in the duodenum where it is regulated by dietary iron (83) D MT1 (-IRE) is expressed in erythroid cell precursors where it is regulated by erythropoietin (95) B esides t heir tissue -specific expression and regulation, these two isoforms may function in different subcellular compartments. In CHO (Chinese hamster ovary) ce lls which stably expressed the DMT1 (-IRE) form, a ring like staining at the periphery of the cells was observed, indicating plasma membrane localization. When cells were p ermeabilized, an intracellular punctate pattern could be seen in addition to membrane staining, indicating its localization in intracellular compartments (96) I n CHO cells, RAW cells (m ouse leukemic monocyte macrop hage cell line) MEL cells ( mouse erythroleukemia cell line) and TM4 cells (mouse sertoli cell line) DMT1 displayed plasma membrane staining and clear colocalization with TF in recycling endosomes (97)

PAGE 35

35 When recom binant DMT1( -IRE) and DMT1( + IRE) were expressed simultaneously in HEp2 cells ( a human larynx carcinoma cell line, which highly express DMT1 (+IRE) isoform ) the two isoforms did not completely colocalize with each other indicating distinct subcellular compartment alization (98) DMT1 ( + IRE) localized in late endosomes and lysosomes in transfected HEp2, HeLa (a human cervical cancer cell line ) and COS 7 cells. In HEp 2 cells, the ear ly endosomal markers EEA1 (early endosomal antigen 1) and TF R 1 were not colocalized with DMT1, but most of the intracellular puncta of DMT1 c ompletely colocalized with LAMP 2 (lysosomeassociated membrane protein 2). Subcellular fractionation by sucr ose gradient revealed that the DMT1 (+ IRE) isoform co -sedimented with the late endosomal and lysosomal membrane proteins LAMP1 (lysosomeassociated membrane protein 1) and LAMP 2, but not with the TF R 1 in early endosomes (99) The a cidic environment in the late endosomal and lysosomal may provide DMT1 the proton gradient required for transport activity DMT1( IRE) colocalized with the early endosomal marker s TF R 1 and EEA1 (98) indicating its localization in early endosomes. In DMT1, t he c arboxy cytoplasmic tail is the only different part of +IRE and IRE forms, indicating the possible existence of targeting signals for different localizat ion in this domain. S ubstitution of the Tyr555 or Leu557 with Ala significantly affected the early endosomal localization of DMT1(IRE) iso form, resulting in its mistargeting t o late endosomes and lysosomes. It was concluded that the Y555XLXX sequence in t he C te rminal cytoplasmic tail of DMT1 (-IRE) is critical for the early endosomal targeting (98) Sequential deletion of almost all the amino acids in the C terminal tail of DMT1(+IRE)

PAGE 36

36 isoform did not affect its late endosomal and lysosomal localizations, indicating that C terminal tail of DMT1(+IRE) does not include the targeting determinant. DMT1 is not the O nly Transporter I nvolved in Endosomal I ron R elease Iron is transported in the circulation between absorption, storage, and utilization sites by TF (100) In human plasma the TF concentration normally range s from 22 to (101) Iron taken up by means of TF R 1 must be transported across the endosomal membrane to be released into the cytosol (d epicted in Fig. 1 4) Based on the obser vation that the loss of TF R 1 produces more severe effects than the loss of DMT1, other transporters functioning in the TF cycle have been hypothesized in animals (86) In b rats, which have a loss of function mutati on in DMT1 TBI uptake was still effe c tive in crypt cells of small intestine, where DMT1 was not detectable (102) Since crypt epithelial cells depend mainly on TF cycle for iron uptake, this indicat e s that DMT1 is not required for the uptake of TBI by duodenal crypt epithelial cells By in activating the mouse DMT1 gene globally and in selected tissues through gene targeting and homologous recombination, Gunshin et al. (86) found that fetal DMT1 was not needed for maternofetal iron transfer but that D MT1 activity was essential for intestinal non heme iron absorption after birth. DMT1 was also required for normal hemoglobin production during the development of erythroid precursors. Iron dextran administered to DMT1/ mice markedly increased liver iron stores in both hepatocytes and macrophages, indicating that hepatocytes and other cells must have an alternative iron uptake mechanism A recent study has shown that expression of MCOLN1 ( mucolipin1 ) which belongs to the mucolipin subfamily of transient re ceptor potential ( TRP) proteins1, mediates iron

PAGE 37

37 release from late endosomes and lysosomes in HEK 293T cells (103) Bargal et al. (104) identified the MCOLN1 gen e which encod es a 580amin o acid protein termed mucolipin1. The MCOLN1 protein contains 1 transmembrane helix in the N -terminal region and at least 5 transmembrane domains in the C -terminal region. Mutations in the human MCOLN1 gene cause Mucolipidosis type IV disease which is an autosomal recessiv e neurodegenerative lysosomal storage disorder characterized by psychomotor retardation, ophthalmologic abnormalities, as well as irondeficiency anemia (105) In summary DMT1 can function in both dietary iron absorption and endo somal iron release. However, DMT1/ mice were born alive with elevated hepatic stores indicating that hepatocytes have an alternative DMT1-independent iron uptake pathway and that DMT1 is not required for maternofetal iron transport, which relies primari ly on TF -bound iron delivery. Moreover, i n Belgrade rats DMT1 is not required for TBI uptake of small intestine crypt epithelial cells. Human cases of DMT1 mutation s also indicate the existence of DMT1-independent iron uptake by both reticulocytes and hep atocytes Therefore, other endosomal transporters must exist for transporting iron from TF into the cytosol Iron U ptake by Hepatocytes The liver plays at least three major roles in iron homeostasis. Firstly, it is the main storage organ for iron. He patocy tes as well as Kupffer cells serve as iron depots, in which excess iron is stored as ferritin and hemosiderin (106) Secondly, the liver regulates iron transfer into and around the b ody through the peptide hormone hepcidin production. Thirdly, the liver is the major site for the synthesis of ironrelated plasma proteins such as TF and c eruloplasmin (Cp) (65) Hepatocytes comprise about 80% of

PAGE 38

38 the total cell s in the liver (107) They can acquire iron through both TBI and NTBI pathways (Fig 1 -7 ). Transferrin -bound Iron Uptake by Hepatocytes In normal situations the majority of circulating ir on is bound to TF T he uptake of TF -bound iron by hepatocytes from plasma is mediated by transferrin receptor In addition, a receptor independent TBI uptake pathway has also been proposed to exist in hepatocytes (10 8,109) By using perfused liver isolated primary hepatocytes or hepatoma cell lines, it has been shown that there is a saturable highaffinity TBI uptake described as a TFR1 dependent pathway and a nonsaturable low affinity process for TBI uptake which i s independent of TFR1 (108,110,111) The uptake of holo -TF through TFR1 is a highly regulated process and most regulation occurs at the level of posttranscriptional mRNA stability. Iron responsive elements (IREs) ar e present in the 3 UTR of TFR1 mRNA (Fig. 1 3). When intracellular iron content is low, IRPs bind to the IRE, protecting the mRNA from endonuclease cleavage. Consequently more TFR1 protein is synthesized. Accordingly, in iron-deficient animals or humans, TFR1 levels are higher in the liver and other tissues. During iron excess, IRPs are inactivated by iron-sulfur cluster formation, thus inhibiting binding to IREs, resulting in increased degradation of TFR1 message and protecting cells from accumulating mor e iron (112) The highaffinity process becomes saturated at relatively low extracellular TF concentrations (50 to 100 nM) (108,109) It i s likely that the receptor independent low affinity pathway predominates at normal physiological level of plasma TF (25 M to 50 M) (113) Under certain pathologi cal circumstances when the iron carrying capacity of TF becomes exceeded, such as hemochromatosis TFR1 i s down-regulated in

PAGE 39

39 hepatocytes. I n untreated hereditary hemochromatosis (H H) patients the expression of TF R1 was undetectable in hepatocytes (114) TFR1 is not the only transferrin receptor in hepatocytes. TF R2 a homolog of TFR1 is highly expressed in hepatocytes and in developing erythroid precursor cells, and it may play a role in liver iron loading (115) Similar to the ubiquitously expressed TFR1, TFR2 is a type-2 membrane protein with a cytoplasmic N -terminus, a single transmembrane domain and a large ectodomain. The amino acid sequences of transmembrane region and the extracellular domain in TF R2 shares 45% identity to that o f TF R 1 T F R2 protein is up-regulated in iron overload and in a mouse model of HH and may contribute to increased TBI uptake by the liver during iron overload conditions (116) Mutations in TF R2 cause hemochromatosis. TF R2 can bind and internalize holoTF, but its affinity for holo -TF is 2 5 -30 times lower than that of TFR1 (117,118) Unlike TFR1, TFR2 lacks an IRE, the protein levels increase in response to increased level of holo -TF (119,120) A study of TFR2 null mice indicated that TFR2 has a minor role in iron transport and hepatic iron-loading (121) Non -transferrin -bound Iron Uptake by Hepatocytes The amount of circulating TF bound iron is determined by three coordinated process: macrophage iron recycling, duodenal iron absorption and hepatic iron storag e /release (122) As TF becomes saturated in iron overload states, excess iron is also found as NTBI NTBI will exist in high amount and contributes considerably to hepatic iron loading (123,124). The form of NTBI present in the plasma could be bound to either citrate or albumin (125) NTB I is likely to play an important role in hepatocyte iron loading in HH and other iron overload conditions. NTBI is cleared rapidly by the liver from plasma, demonstrated by cases of human and mouse congenital TF

PAGE 40

40 deficiency (atransferrinemia and h ypotransfe rrinemia) (51,126,127) Affected individuals absorb dietary iron efficiently and deposit large amounts in the liver even though they lack TF. It has been shown that after TF saturation, 70% of the NTBI given orally or intravenously, was taken by the liver (128) The molecular basis of NTBI uptake is not well understood In summa ry, t he live r can take up both TBI and NTBI. DMT1 the only known iron transporter involved in TBI release from endosomes may also mediate NTBI into cells. However, DMT1 knockout mice can still accumulate iron in the liver, indicating that DMT1 is not the major iron transporter in either pathway for this organ. How iron is taken up into the liver, especially into hepatocytes, remains to be elucidated (9) ZIP14 and Iron Metabolism Identification and Characterization of ZIP14 ZIP14 (SLC39A14, solute carrier 39 family, member 14), as one of the members of ZIP metal -ion transport er superfamily can transport both zinc and iron (129) It is the second identified iron-import protein to date (the first one is DMT1). The ZIP family protein takes the name from Z RT, I RT like p rotein, where ZRT represents z inc r egulated t ransporter and IRT stands for i ron-r egulated t ransporter (130) By sequencing clones obtained from the cDNA library of a human immature myeloid cell line (KG -1 cells) Nomura et al. (131) first cloned ZIP14 which they named KIAA0062. Northern blot analysis detected that ZIP1 4 was ubiquitously expressed, with h ighest expression in the liver, and lowest expression was in the spleen, thymus, and peripheral blood leukocytes. A m ultiple tissue expression array also showed ubiquitous expression of ZIP14 with high expression in the liver, pancreas and heart (132) By searching databases for sequences that were similar to a unique HEXPHEXGD motif (a

PAGE 41

41 consensus sequence for the catalytic zinc binding site of metalloproteases within ZIP proteins), Taylor et al. (133) identified ZIP14 which they designated LZT -Hs4. Human Zip14 was predicted to be 53 kDa with up to three additional N -linked carbohydrate side cha ins. Expression of human ZIP14 in CHO cells showed an apparent molecular mass of 53 kDa Under nonreducing conditions, ZIP14 migrated as a trimer. ZIP 14 was expressed on the plasma membrane in nonpermeabili z ed cells Human ZIP14 was predicted to have a lon g N terminus, followed by 8 putative transmembrane domains and a short C terminus (129,132,133) The metal transport activity of ZIP14 was first characterized in 2005 by Taylor et al. (132) who showed that ZIP14 overexpression in CHO cells stimulated the u ptake of zinc into the cytosol. In 2005, Liuzzi et al. (134) found that t ransfection of mouse Zip14 cDNA into HEK 293 cells increas ed zinc uptake and that Zip14 was the most upregulated zinc transporter in response to lipopolysaccharide (LPS) treatment or turpentine induced inflammation in the mouse liver. However, the hypozinc emic response was milder in interleukin6 knockout (IL -6/ -) mice exposed to LPS than in wildtype mice. IL 6/ mice display ed neither hypozincemia nor Zip14 induction with turpentine induced inflammation Immunohistochemical analysis showed that in hepatocytes, plasma membrane expression of Zip14 i ncreased in response to both LPS and turpentine. IL -6 also increased expression of Zip14 in primary hepatocyte cultures. It was concluded that ZIP14 is a zinc importer upregulated by IL6 and Zip14 plays a major role in the hypozincemia accompanying the acute-phase re sponse to inflammation and infection.

PAGE 42

42 ZIP14 and Cellular Iron Uptake I n 2006, Liuzzi et al. (129) reported that overexpres sion of Zip14 in HEK 293 cells and Sf9 insect cells enhanced not only the uptake of zinc, but also iron. The iron was presented to the cells as ferric citrate, the major form of NTBI that appears in plasma in conditions of iron overload (135) The iron uptake was inhibited by zinc suggesting that these two metals share the same transporter It was also shown that siRNA -mediated suppression of endogenous Z ip 14 expression in AML12 cells, a mouse hepatocyte cell line, resu lted in reduced uptak e of NTBI from ferric citrate. In 2008, Gao et al. (136) showed that ZIP14 overexpression stimulated NTBI uptake in HeL a cells, and that suppression of endogenous ZIP14 in the human hepatoma cel l line, HepG2, decreased NTBI uptake. Moreover, Zip14 can efficiently transport iron at pH 7.4 (129) the pH at the plasma membrane surface of hepatocytes In contrast DMT1, the first identified iron importer transports iron optimally at pH 5.5 (18,137) and is readily detected in endosomes but not on the plasma membrane of hepatocytes (138) These observations provide strong support that ZIP14, rather than DMT1 predominantly mediates NTBI uptake into h epatocytes Under normal conditions, the majority of iron in the plasma is bound to T F. When t he amount of iron in the plasma increase s and exceeds the capacity of TF, the level of NTBI increases, contribut ing to t issue iron loading. In iron-loading disor ders, such as hemochromatosis or dietary iron overload, excess iron is mainly found in the liver, pancreas and heart (2) HFE associated HH is the most prevalent form of hemochromatosis and is one of the most common inherited human disorders (139) In this disease, m utation of a single amino acid (C282Y) in the hereditary hemochromatosis gene ( H FE ) causes iron overload (140) The functional role of HFE in

PAGE 43

43 iron metabolism is also supported by the evidence that hepatocytes from Hfe/ mice can take up more NTBI compared to wide -type mice (141) However, the mechanism by which HFE regulates iron metabolism still remains to be determined Interestingly, overexpression of HFE in HepG2 cells decreased ZIP14 levels by decreasing the stability of ZIP14 (136) T he reduced ZIP14 levels were associated with diminished uptake of not only NTBI but also TBI suggesting that ZIP14 participates in both pathways of iron acquisition A study in isolated rat hepatocytes has shown that LPS markedly increas es the uptake of TBI (142) Similarly in HepG2 cells, stimulation with the inflammatory cytokine, IL -6, enhanced the up take of TBI by 48% (143) LPS and IL6 have both been shown to potent ial ly increase levels of Zip14 in mouse liver and in isolated hepatocytes (134) These observations suggest that ZIP14 is involved in TBI uptake b y hepatocytes. In summary, ZIP14 is a newly identified iron import protein. It localizes to the plasma membrane and mediates the uptake of NTBI into cells. Expression of Zip14 was induced by both LPS and IL-6, which could also stimulate the TBI uptake in hepatocytes. HFE expression decreased ZIP14 stability and reduced the uptake of both NTBI and TBI in HepG2 cells. The above evidence suggests that ZIP14 is a common transporter shared by both NTBI and TBI. The research described herein tested the hypothesi s that ZIP14 plays a role in the uptake of TBI Studies were also directed to investigate the membrane topology of ZIP14 a s a first step to understand which structural elements contribute to the iron transport ability of ZIP14.

PAGE 44

44 Figure 1 1. D istributio n of iron within the body. N ormal adults typically have 35 g of total body iron. To maintain iron bal ance, about 1 2 mg of dietary iron is absorbed every day to replace the iron that is lost in the urine, feces, sweat and sloughed cells. Most body iron ca n be found in mature erythrocytes and in erythroid bone marrow Reticuloendothelial macrophages recycle iron from old red cells and supply mos t of the iron for new red blood cell synthesis. Approximately 0.1% of body iron is bound to TF, a circula ting plas ma protein that deliver s iron to erythroid precursors and other tissues. I ron is stored primarily in hepatocytes.

PAGE 45

45 Figure 1 2. Schematic model for intestinal iron absorption. Dietary non-heme iron (mostly Fe3+) is reduced by a ferric reductase (likely D cytb ) to yield ferrous iron (Fe2+), which subsequently enters the enterocytes via DMT1. Heme is absorbed intact perhaps via h eme carrier protein 1 (HCP1 ) and is then catabolized by heme oxygenase 1 (HO) to li berate Fe2+. If body iron stores are high, iron may be stored in ferritin and lost through sloughed epithelial cells within 3 4 days. Iron efflux across the basolateral membrane into circulation is through FPN 1 The exiting iron is oxidized to Fe3+ by h e phaestin to enable loading onto TF.

PAGE 46

46 Figure 1 3 Overview of the iron-responsive element/iron-regulatory protein (IRE/IRP) network. IRPs interact with IREs to coordinate the expression of proteins involved in iron metabolism. IRP binding to IREs located at 5 untranslated regions (UTRs) inhibits translation, whereas IRP binding to 3 UTR IREs increases mRNA stability. Cellular iron loading turns IRP1 from its IRE binding form to the Fe S cluster -containing inactive form. Low iron levels promote accumulation of active IRP1, resulting in its binding to IREs.

PAGE 47

47 Figure 14. Overview of the TF cycle. When diferric TF binds TFR1 on the cell surface the complex internalizes through receptor mediated endocytosis. Endosomes become acidified by a proton pump. Acidification leads to protein co nformational chang es that cause iron to dissociate from TF. Fe3+ released from TF is reduced by ferrireductase STEAP3, and is then transported out of the endosome, presumably through DMT1. TF and TFR1 both return to the cell surface, where they separate at neutral pH. Both proteins participate in further rounds of iron delivery.

PAGE 48

48 Figure 1 5 Overview of the receptor -mediated pathway for endocytosis of extracellular heme and hemoglobin. CD163 and LRP/CD91 represent two pathways for uptake of extracellular heme incorporated in haptoglobinhemoglobin (Hb-Hp) and hemehemopexin (Heme-Hx) complexes. Both receptors are highly expressed in phagocytic macrophages, which can metabolize heme into bilirubin, Fe and carbon monoxide. In addition to the expression in macrophages, LRP/CD 91 is also highly expressed in several other c ell types including hepatocytes and neurons.

PAGE 49

49 Figure 16 Schematic demonstration of mutations in the iron transport protein DMT1. Identified DMT1 mutations in rodents ( G185R) and human ( V114 del, G212V, E399D and R416C).

PAGE 50

50 Figure 1 7 Iron uptake by hepatocytes. Hepatocytes have several pathways for iron uptake from the circulation including the uptake of TBI (Fe2-TF, diferric TF ) via TFR1, TFR2, and TFR -independent mechanisms. The TFR1 pathway is w ell characterized. When diferric TF binds TFR1, the complex is internalized by endocytosis, and iron is dissociated from T F during endosomal acidification Released iron is transferred to the cytoplasm for ironrelated biological functions or storage as fe rritin. The resulting apo TF -T F R1 complex then returns to the cell surface for reutilization. The detailed pathways through TFR2 and the one independent of TFR1 and TFR2 remain to be elucidated, but are considered to be important for iron uptake in hepatocy tes. The uptake of NTBI, which is present in the plasma during conditions of iron overload, is likely mediated by Z ip 14 and /or DMT1

PAGE 51

51 C HAPTER 2 MATERIALS AND METHOD S Cell C ulture All cells were maintained in an incubator at 37C and 5% CO2. HEK 293T cells were grown in Dulbeccos Modifi ed of Eagles Medium (DMEM, Mediatech) with 4.5 g/L glucose, 4 mM L -glutamine, 1 mM sodium pyruvate, 100 U/ml peni cillin, 100 g /ml streptomycin and 10% fetal bovine serum (FBS, Atlanta Biologicals) HepG2 cells were maintain ed in DMEM with 4.5 g/L glucose, 4 mM L -glutamine, 1 mM sodium pyruvate, 1 MEM nonessential amino acids (Mediatech), 100 U/ml peni cillin, 100 g /ml streptomycin and 10% FBS. Expression of M ouse Zip14 (mZip14) in HEK 293T and HepG2 cells Effectene reagent (Qiagen) was used for tr ansient transfection of HEK 293T cells. Briefly, HEK 293 T cells were seeded at 40% confluency in 6well plates. Transfection began 24 h after seeding with 0.4 g of plasmid DNA, 3.2 l of enhancer and 10 l of Effectene reagent, and was carried out for 48 h before further analysis. For HepG2 cell transfection, JetPei -Hepatocyte transfection reagent (Genesee Scientific) was used. HepG2 cells were seeded at 40% confluency in 6well plates. Twenty -four hours later, 3 g of plasmid DNA, 9.6 l of JetPei -Hepatocyte were diluted in 100 l of 150mM NaCl solution and mixed together. The mixture was incubated at room temperature for 15 min before added to each well. Knockdown of Endogenous ZIP 14 in HepG2 C ells U sing s iRNA SMARTp ool s iRNA (sma ll interfering RNA) specific for human ZIP 14 (Genbank accession no. NM_015359) and non -t argeting pool negative control s iRNA were purchased from Dharmacon (Thermo Scientific). Lipofectamine RNAiMAX transfection

PAGE 52

52 reagent (Invitrogen) was used to transfect si RNA into HepG2 cells. Briefly, 6 l of lipofectamine RNAiMAX and 60 pmol of RNAi duplex were mixed in 600 l Opti MEM medium and added into each well of a 6well plate. After incubati ng at room temperature for 15 min, ~2 105 cells were added to each well in 1.8 ml of DMEM supplemented with 4.5 g/L glucose, 4mM L-glutamine, 1mM sodium pyruvate, 1 MEM nonessential amino acids and 10% FBS. Seventy -two hours later, protein levels were detected by using W estern blot to confirm ZIP 14 knockdown. Measurement of T F -bound Iron U ptake 59Fe -T F was prepared by saturating human apo-TF (Sigma) using 59Fe -ferric nitrilotriacetic acid (NTA). After incubating for 1 h in PBS containing 10 mM NaHCO3, unbound radiolabel ed iron was removed by repeated washing through a Centr icon centrifugal filter (Millpore, MWCO 30,000). HEK 293T or HepG2 cells were washed twice with serum -fr ee DMEM (SFM) and incubated at 37C for 60 min to deplete T F from cells. Cells were then inc ubated in SFM containing 100 nM 59Fe -TF, 20 mM H EPES (pH 7.4) and 2 mg/ml ovalbumin. To stop uptake, cells were placed on ice, and externally bound 59Fe -TF was stripped with an acidic buffer (0.2 N acetic acid, 500 mM NaCl, 1 mM FeCl3) for 3 min. After washing suspended cells twice in cold SFM, cells were lysed in a solution containing 0.2% SDS and 0.2 M NaOH. Cell associated radioactivity was determined -counting. Uptake was expressed as cpm/mg protein. pH D ependence of Iron Transport A ctivity The pH of the uptake buffer (130 mM NaCl, 10 mM KCl, 1 mM CaCl2 and 1 mM MgSO4) was adjusted from pH 7.5, 6.5, to 5.5 by using mixtures of HEPES and MES buff er s (144) NTBI uptake was measured as described previously (129) Briefl y, transfected HEK 293 T cells were washed three times in SFM and then incubated for 1 h

PAGE 53

53 in SFM to deplete cells of TF For uptake, cells were incubated with 2 M 59Fe -ferric citrate for 60 min, followed by three washes of cell impermeant iron chelator solution to remove surface -bound iron. Cell associated radioactivity was determined by using a counter Uptake was expressed as c pm/mg protein. Genetic Knockin to Tag E ndogenous Z IP 1 4 of HepG2 C ells with 3Flag Epitope AAV knock -in vector construction The epitope tagging st rategy and method have been described in detail previously (145,146) Briefly, about 1 kb of left and right homologous arms of ZIP14 were PCR amplified by using Platinum Taq DNA polymerase High Fidelity (Invitrogen) and genomic DNA isolated from HepG2 cells The primers for the knockin (Table 2 -1) were designed specifica lly for sequences upstream or downstream of the stop codon. Both arms were inserted into the rAAV-NeoLox P 3Flag knock in vector (Kindly provided by Dr. Zhenghe Wang Case Western Reserve University ) by using the USERTM enzyme system (NEB). The entire US ER -treated reaction mixture was used to transform chemically competent E. coli (Escherichia coli ) by heat shock at 42C. AAV Cre expression vector construction To make the AAVCre encoding v ector the C re gene coding sequence was PCR amplified from pBS185 plasmid (Addgene) by using gene -specific primers linked with Not I restriction site s at each end (Table 2 1), and then ligated into pAAV vector (Stratagene). Packaging of AAV viruses Targeting viruses were packaged into HEK 293AAV cells by using AAV H elper -Free System (Stratagene) according to the manufacturers instructions V iruses were harvested after 72h. Infection of HepG2 cells with AAV virus and Screening for targeting clones HepG2 cells growing in T25 flasks were infected with targeting virus and G418

PAGE 54

54 resistant single -cell clones were transferred to 24well plates for screening and expanding. After confirming homologous recombination by PCR with screening primers (Table 2 1) and by DNA sequencing of the PCR product the positive clones were i nfected with AAV -C re virus to remove the Neomycin-resistant (NEOR) marker Successful excision was confirmed by PCR with screening primers Expression of endogenous Zip143Flag i n HepG2 cells was confirmed by W estern blo t t ing Western B lot Analysis Cells were washed with old phosphatebuffered saline (PBS) twice and lysed in SDS lysis buffer [ 170mM Tris -HCL (pH 6.8) mercaptoethanol, and 1 Complete M ini protease inhibitor Mixture (Roche)] Protein concentrations of the cell lysa te s were measured by using the RC DC Protein Assay (Bio -Rad). S amples were mixed with 1 Laemmli buffer and incubated for 15 min at 37C. Proteins were separated electrophoretically on an SDS/7.5% or 10% polyacrylamide gel, transferred to nitrocellulose and incubated for 1 h in blocking buffer (5% nonfat dry milk in Tris -buffered saline-Tween 20, TBST). Blots were incubated for 1 h at room temperature in blocking buffer containing 2.5 g/ml affinity purified rabbit anti Zip14 antibod y, mouse anti Flag, M2 (1:2000, Sigma), mouse anti -T F R 1 /TFR2 (1:1000, invitrogen) or rabbit anti -DMT1 (1:500, Abcam). After four washes with TBST, blots were incubated with a 1:2000 dilution of donkey anti -rabbit (Amersham Biosciences) or goat anti mouse (Invitrogen) secondary ant ibodies conjugated to horseradish peroxidase (HRP). To confirm equivalent loading, blots were stripped for 15 min in Restore PLUS Western Blot Stripping Buffer (Thermo Scientific) blocked for 1 h in blocking buffer, and reprobed with rabbit anti actin (1: 1000, Sigma) followed by HRP conjugated donkey anti -rabbit secondary antibody. After two washes with TBST and

PAGE 55

55 TBS, antibody cross-reactivity was visualized by using enhanced chemiluminescence (SuperSignal West Pico, Thermo Scientific ) and x -ray film. Subce llular L ocalization Cells seeded on poly -L lysine (Sigma) coated coverslips were washed two times with PBS+/+ (PBS with 1mM MgCl2 and 0.1mM CaCl2) and fixed with 2% paraformaldehyde (PFA) for 15 min at room temperature. Next, cells were washed three times with PBS, permeab i lized with 0.1% saponin for 10 min, and then washed three times with PBS before block ing in 1% bovine serum albumin (BSA) for 30 min. For T F labeling, cells were incubated with 30 g /ml Alexafluor 488-labeled human holo -T F (Invitrogen) f or 30 min at 37C and washed three times with PBS before fixation. To stain the nuclei, cells were washed three times with PBS and incubated for 5 min with 10 g/ml 4' -6 diamidino 2 -phenylindole (DAPI). After several washes of PBS, coverslips were mounted on microscope slides with mounting medium ( Vector Laboratories) and sealed with nail polish. Images were captured with a Leica TCS SP5 laser -scanning confocal microscope. Quantitative Determination of TF u ptake To determine if ZIP14 knockdown affected the uptake of TF, HepG2 cells were incubated with 100 nM biotin-labeled holo -TF (Sigma) After 4 h, uptake and surface binding of TF were stopped by trypsinization. After washing suspended cells twice in cold SFM, cells were lysed in SDS lysis buffer. Cell ex tracts were analyzed by Western blotting as described above, but with ImmunoPure HRP -conjugated streptavidin (1:50,000 for 1 h) instead of antibodies.

PAGE 56

56 Isolation of P lasma M embrane P roteins by C ell-s urface B iotinylation HepG2 cells were incubated overnight in SFM with 10 M apo -TF or 10 M holo TF. For wi l d -type mZip14 and the asparagine mutants the transporters were expressed in HEK293T cells by transient transfection as described above. Twenty -four h ours after transfection, the medium was removed, and t he cells were washed twice with ice -cold PBS. Plasma membrane proteins were isolated by using the Cell Surface Protein Isolation Kit (Thermo Scientific) according to the manufacturers protocol. Briefly, the flasks were kept on ice, and all solutions were ice -cold for the rest of the procedure. Each flask of cells was incubated with 10 ml of the membrane-impermeant biotinylation reagent, NHS -SS -biotin (0.25mg/ml in PBS) for 30 min with very gentle shaking. After biotinylation, each flask was added 500 l qu enching solution to quench the unreacted NHS -SSbiotin. Cells were collected and lysed in 500 l lysis buffer with 1 protease inhibitors ( Roche) followed by centrifugation at 10,000 g for 2 min at 4C The clarified supernatant was added to a spin column containing pre washed immobilized NeutrAvidin gel and incubated for 60 min at room temperature. After four washes biotinylated samples were incubated with 50 mM DTT in 1 SDS PAGE sample buffer for 60 min at room temperature to cleave the disulfide bond and release biotinylated proteins. Cell -surface expression of ZIP14, TFR2, Lamp1 tubulin, and Na+, K+ ATPase were determined by Western blotting as above, but wi th these additional antibodies: mouse anti Lamp1 or Na+, K+ ATPase ( 1:1000, Santa Cruz) and mouse anti tubulin, clone B 5 -1 -2 (1:5000, Sigma) RNA I solation and Q uantitative PCR (qPCR) Tot al RNA was isolated from HepG2 and HEK 293T cells by using RNABee (TelTest) Isolated RNA was treated with DNaseI (Turbo DNA -free kit, Ambion) to

PAGE 57

57 remove any cont aminating genomic DNA. First -strand cDNA was synthesized from the isolated RNA by using the High-Capacity cDNA Archive kit (Applied Biosystems). Quantitative RT -PCR was performed using iQ SYBRGreen Supermix (Bio-Rad) and an Applied Biosystems 7300 realtime PCR system. Copy numbers of DMT1 and ZIP14 mRNA were calculated by comparing Ct values obtained from HEK 293T and HepG2 RNA to those obtained from standard curves generated by using the plasmids pBluescriptR human DMT1 (BC100014, Addgene) and pXL4-human Z IP14 (NM_001135153.1, Open Biosystems). The primers used for ZIP14 (forward: 5' CTGGACCACATGATTCCTCAG -3; reverse: 5 -GAGTAGCGGACACCTTTCAG -3) and DMT1 (5' -TGGTTCTGACTCGCTCTATTGC -3; reverse: 5 CATTCATCCCTGTTAGATGCTCTACA -3) were designed to target all kn own variants of ZIP14 and DMT1 mRNA. Construction of GFP F usion and Flag-tagged mZip14 Plasmids An expression vector for C -terminal mZip14 -EGFP was constructed by amplifying mouse Zip14 coding sequence (BC021530) from pCMV -Sport6 mZip14 vector (Open Biosy stems ) and ligating into pEGFP -N1 (Kindly provided by Dr Ivana DeDomenico, University of Utah ). To make a C -terminal mZip14 -3 Flag vector (C T -Flag mZip14) the 3 Flag sequence followed by a stop codon was amplified from pCMV 3tag3a (Stratagene) by a f orward primer linked with a Sal I site and a reverse primer linked with Not I site ( P1 +P2, Table 2 -2 ). T he PCR product was ligated into mZip14 -EGFP vector after excision of the EGFP sequence by Sal I and Not I restriction enzyme s To insert the 3 Flag sequence into the N termin us and the histidine -rich region loop (N -T -FlagmZip14 and His -FlagmZip14), site directed mutagenesis (method see below) was first used to add an Spe I restriction site in the N -terminus (after the predicted signal peptide

PAGE 58

58 cleavage site) and in the histi di ne -rich loop (immediately after the histi di ne 254 in the HHHGHNH motif ) of the pCMV -Sport6 mZip14 vector (b y primers P3+P4 or P5+P6, Table 2 2 ). The 3 Flag sequence was amplified fr om pCMV -3tag-3a with Spe I site s added at each end (By primers P7+P8, Table 2 -2 ), followed by ligating the PCR product into the Spe I siteadded vectors. All constructs were verified by DNA sequencing. Primary Sequence Analysis and Bioinformatic P rediction s The m ouse Zip14 gene sequence ( NC_000080.5) an d its protein sequences, isoform a ( NP_001128623.1) and isoform b ( NP_659057.2) were obtained from the NCBI website. Sequence alignment and consensus analysis were performed by using Vector NTI 11.0 software (Invitrogen). Prediction of the transmembrane t opology and signal peptide sequence were performed by using the following online programs: TMHMM (147) HMMTOP (148) TopPred (149) MINNOU (150) TmPred (151) ConPred II (152) MEMSAT -SVM & MEMSAT III (153) Phobius (154) SOSUI (155) SignalP3.0 & SignalP V2.0b2 (156) SP EPlip (157) PrediSi (158) and SIG Pred (bmbpcu36. leeds.ac.uk/prot_analysis/Signal.html ). Prediction of protein glycosylation status was done by using the online programs as follows: NetNGlyc 1.0 (http://www.cbs.dtu.dk/services/NetNGlyc/) and NetOGlyc 3.1 (159) Construction of Asparagine Mutant Plasmids Mouse Zip14 contains five predicted N -lin k ed glycosylation sites as asparagines 52, 75 85, 100 and 455. The codons for each asparagine ( N ), were mutated to encode for aspartic acid ( D ), giving different N to -D mutants. Site directed mutagenesis was performed by using the Quickchang e II Kit (Stratag ene). In brief, 100 ng of doublestranded DNA template ( N -T -FlagmZip14 or C -T -FlagmZip14 ) was mixed with the primer s (f orward and reverse primers, 125 ng each, Table 2 3 ), 10 mM dNTPs, 1

PAGE 59

59 reaction buffer, and Pfu DNA polymerase. The mixture was amplified by polymerase chain reaction. Initially the reaction mix was incubated at 95 C for 30 s. Cycles were as follows: denaturation for 30 s at 95 C, annealing for 1 min at 55 C, and extension synthesis at 68 C for 7 min for 1 8 cycles. PCR products were dige sted with DpnI enzyme to remove the parental strands. The digested DNA mixture was transformed into E. coli XL1blue cells by heat shock at 42C Mutagene sis products were all verified by DNA sequencing. Measurement of Iron Transport A ctivity NTBI uptake was measured as previously described (129) Briefly, transfected HEK 293T cells were washed three times in SFM and then incubated for 2 h in SFM to deplete cells of TF For uptake, cells were incubated with 2 M 59F e -ferric citrate for 2 h, followed by three washes of cell -impermeant iron chelator solution to remove surfacebound iron. Cell -counter. Uptake was expressed as cpm/mg protein. Statistical A nalysis Data represent mean SEM. The TBI uptake studies were analyzed by unpaired Students t -test (GraphPad Prism, GraphPad Software ). Iron transport activity studies of the asparagine mutants were analyzed by oneway analysis of variance and Tukeys post hoc test (GraphPad Prism, GraphPad Software) The pH -dependence of iron uptake data were analyzed as a completely randomized block design using the Glimmix procedur es (SAS Inst. Inc.). B locks were based on the replication of the experiment. Fixed effect included treatment and pH as well as the interactive effects of treatment and pH. Block was the random effect. Multiple compa rison were adjusted by Tukey -

PAGE 60

60 Kramer. Degrees of freedom were approximated using the KenwardRogers method (DDFM = kr). A probability level of P < 0.05 was defined as a significant difference.

PAGE 61

61 Table 2 1. List of Primers Used for PCR to G enerate and Screen for E ndogenous ZIP14 3 Flag in HepG2 C ells Primer S et Forward P rimer Reverse P rimer K nock in Primers Left arm 5' GGGAAAGUGCAACCTT 5 GGAGACAUTCCCCAATC GAACTCCGGAGC 3 TGGATCTGTCCTGAAT 3 Right arm 5 GGTCCCAUGGCTCTGC 5' GGCATAGUGCACCCCAT CAAGAGCCTG 3 TTCTACAAGTCAGC 3 Cre specific 5' ATATTGCGGCCGCAAG C 5' AATAAGCGGCCGCCGCG primers TTGGCCCATTGCATAC 3 TTAATGGCTAATCGCC 3 Screening P1: 5 CCATTCAGCGGTTTTT P2: 5' GTTGTG CCCAGTCAT primers AAGGGGGC 3 AGCCG 3 P3: 5' GGATTCATCGACTGT P4: 5' CAAGGGCTCCACAGT GGCCG 3 GGCTAAG 3 P5: 5' GGCCTCCTGACTGGA P6: 5' GAGACTGGTTACCAG TTCACC 3 G GCAGC 3 Table 2 2. List of Primers Used for G enerating 3 Flag -tagged mZip14 V ectors Primer Name Primer Sequence 3 Flag Forward P1: 5' ATCGATACCGTCGACCTCGAG 3' (Sal I + Not I) Reverse P2: 5' AATAAGCGGCCGCCTATTTATCGTCAT CAT 3' N T Spe I Forward P3: 5' CTGCCGCCCCTCACTAGTGCCACCTCC 3' Reverse P4: 5' GGAGGTGGCACTAGTGAGGGGCGGCAG 3' His Spe I Forward P5: 5' ATCACGGGCATAACCATACTAGTTTTACCTCCG AGACACT 3' Reverse P6: 5' AGTGTCTCGGAGGTAAAACTAGTATG GTTATGC CCGTGAT 3' 3 Flag Spe I Forward P7: 5 ATATTACTAGTCAGATTACAAGGATGACGACGA TA 3' Reverse P8: 5' AATAAACTAGTTTTATCGTCATCATCTTTGTAGT CC 3'

PAGE 62

62 Table 2 3. List of Primers Used for Generating A sparagine M utants Mutant Name Primer Sequence N52D Forward 5' GGACCGCTATGGAAAGGATGACAGCCTTAC CCT 3' Reverse 5' AGGGTAAGGCTGTCATCCTTTCCATAGCGGT CC 3' N75D Forward 5' GAGTGGGCCGGGATGATGTTTCCCAGCC 3' Reverse 5' GGCTGGGAA ACATCATCCCGGCCCACTC 3' N85D Forward 5' GGAAGGACCCAGGGACCTCTCCACGTG 3' Reverse 5' CACGTGGAGAGGTCCCTGGGTCCTTCC 3' N100D Forward 5' CTTTGCGGCGCACGACTTGAGCGAGCG 3' Reverse 5' CGCTCGCTCAAGTCGTGCGCCGCAAAG 3' N455D Forward 5' CCAGGAGGAT GAGAAGGACGACAGCTTTCTG GT 3' Reverse 5' ACCAGAAAGCTGTCGTCCTTCTCATCCTCCT GG 3'

PAGE 63

63 CHAPTER 3 ZIP14 AND TRANSFERRI N -BOUND IRON UPTAKE Introduction Iron is essential for almost all known organisms. Iron uptake by cells is a carrier m ediated process and is primarily through the TF -TFR1 complex, a process known as TBI uptake. This receptor mediated endocytosis process involves the transfer of iron out of endosomes into the cytosol Normally, iron is transported in the plasma bound to TF But w hen the iron carrying capacity of TF becomes exceeded during conditions of iron overload, such as in hemochromatosis and thalassemia, NTBI may present in high q u antities. Thus, TBI and NTBI will coexist in the plasma (160,161) Under normal conditions, the liver takes up TBI, almost exclusively into hepatocytes (162) The uptake of NTBI into hepatocyte cell lines is mediated, at least in part, by the transmembrane protein ZIP14, a member of the ZIP family of metal -ion transporters (129,136) At least three studies have shown that TBI and NTBI compete for uptake by hepatocytes (160,163,164) suggestin g the existence of transporter(s) shared by these two pathways. Up to now, DMT1 a major ferrous transporter for dietary absorption, is the only known transport protein which can function in endos o mal iron release. However, at least three lines of evidence support the existence of other iron transporter(s) involved in TBI uptake by cells. Firstly, DMT1 knockout ( DMT1/ -) mice were born alive (86) indicating fetal DMT1 is not required for maternofetal iron transport, which relies primarily on TBI delivery (165,166) Secondly, in Belgrade rats, which have a loss of -function mutation in DMT1 TBI uptake was still effective in duodenal crypt cells which depend mainly on the TF cy cle for iron uptake, indicating that DMT1 is not required for the uptake of TBI by these cells (102) Thirdly, in the case

PAGE 64

64 of DMT1 mutation s in humans severe liver iron loading occurred after blood transfusion and administration of rEpo resulted in rapid improvement of anemia, suggesting t he existence of a DMT1-independent pathway for iron utilization by TBI in reticulocytes and iron delivery by TBI and/or NTBI in the liver (93) Su rprisingly, overexpression of HFE reduced both TBI uptake and NTBI uptake in HepG2 cells (136) .The e ffect of HFE on NTBI uptake was abolished by Z IP 14 knockdown in these cells, confirming the role of Z IP 14 in NTBI uptake. Lack of functional hereditary hemochromatosis protein, HFE, causes iron overload in the liver, heart and pancreas, mostly in hepatocytes, which are the major site s for HFE expression in the liver (167) The strongest evidence that suggests Z IP 14 is involved in TBI uptake is that overexpression of HFE decreased Z IP 14 protein level s by promoting Z IP 14 degradation. The reduced ZIP14 levels were associated with diminished uptake of not only NTBI but also TBI in HepG2 cells. However whether or not Z IP 14 is involved in TBI uptake remains to be determined In this chapter the function of Z IP 14 in TBI uptake was studied in HEK 293T and HepG2 cells. The subcellular localization of Z ip 14 was studied in HepG2 cells. By overexpressing mZip14 in HEK293T cells, I not only found about 25% increased TBI uptake, but also found that Zip14 can transport iron at pH 6.5, suggesting that ZIP14 co uld be a candidate for endosomal iron release. Importantly, siRNA mediated k nockdown of endogenous ZIP 14 decreased TBI uptake in HepG2 cells by 45%. Consistent with a role in TBI uptake, ZIP14 was found to localize not only at the plasma membrane, but also with endocytosed TF. Taken together, t hese results suggest that Z IP 14 may play a role in TBI uptake in hepatocytes.

PAGE 65

65 Result s Overexpression of Zip14 Increases TBI Uptake in HEK 293T C ells HEK 293T cells, an easily transfectable cell line, were used to inv estigate the effect of Zip14 overexpression on TBI uptake. I found that HEK 293T cells transfected with mZip14 took up 25% more 59Fe -TF than did control cells transfected with empty vector (Fig. 3 -1A). Western blot analysis confirmed that the enhanced uptake of 59Fe -TF was associated with higher Zip14 protein levels (Fig. 3 1B). I also measured levels of TFR1 and DMT1, which may also function in this iron uptake pathway. Levels of these proteins did not differ between cells overexpressing Zip14 and controls (Fig. 3 -1B). These results suggest that Zip14 may play a role in TBI uptake. pH -dependent Iron Transport A ctivity of Zip14 To determine the pH dependence of Zip14mediated iron transport, I transfected HEK 293T cells with Zip14 or empty vector, and measured the uptake of 59Fe -ferric citrate by cells incubated in medium at pH 7.5, 6.5, or 5.5. I found that cells overexpressing Zip14 took up more iron than did controls at pH 7.5 and 6.5, but not at 5.5 (Fig. 3 2 ). Subcellular L ocalization of Zip14 -GFP in He pG2 C ells I used confocal laser -scanning microscopy to examine the subcellular distribution of Zip14 -GFP in HepG2 cells after transient expression Zip14-GFP was readily detectable at the hepatocyte plasma membrane and displayed abundant punctuate intracel lular staining that partially colocalized with internalized Texas red labeled TF a marker of recycling endosomes (Fig. 3-3). Localization of Zip14 -GFP to the plasma membrane is consistent with its postulated role in the uptake of NTBI at the cell surface.

PAGE 66

66 Detection of Zip14 -GFP in recycling endosomes further implicates this protein in the assimilation of iron from TF ZIP14 and DMT1 mRNA Copy Numbers in HepG2 and HEK 293T C ells ZIP14 is most abundantly expressed in the liver (131,132) a major organ of iron metabolism. It has been reported that ZIP14 mRNA levels (relative to GAPDH mRNA) were approximately 6-fold higher than DMT1 mRNA levels in HepG2 cells (136) Here I measur ed absolute mRNA copy numbers of ZIP14 in HepG2 and HEK 293T cells, and compared them to DMT1 mRNA copy numbers. In HepG2 cells, ZIP14 mRNA copy numbers were found to be 11-fold greater than DMT1 copy numbers (Fig. 3-4 ). Moreover, ZIP14 transcript levels i n HepG2 cells were found to be 11 -fold greater than in HEK 293T cells. Likewise, in human tissues ZIP14 is much more abundantly expressed in the liver than in the kidney (132) The high expression level of ZIP14 in the liver and HepG2 cells suggests that ZIP14 plays an important role in hepatocytes. Epitope tagging of Human Endogenous ZIP14 in HepG2 C ells To investigate the role and subcellular localization of ZIP14 in HepG2 cells, I first generated a cell line that expresses Flag -tagged ZI P14 from its endogenous locus. The addition of a Flag-tag to ZIP14 allows for highly specific and sensitive detection of the endogenous protein by the mouse anti -Flag M2 monoclonal antibody. Flag -tag encoding DNA was knocked into th e endogenous ZIP14 locus by homologous recombination (Fig. 3 5 A). Correctly targeted neomycinresistant ( NeoR) clones were identified by genomic PCR. Primers used for the knock -in and screening are shown in Table 2 1. Screening primers were designed to reg ions of the Flag + NeoR allele predicted to be ~1.8 Kbp (P1 + P2) and ~1.6 Kbp (P3 + P4). Fig. 4B shows the PCR identification of a correctly targe ted clone. DNA sequencing of the PCR products

PAGE 67

67 confirmed correct knock -in of the flag epitope at the C -terminus and insertion of the NeoR gene. Excision of the NeoR gene and identification of a Flag allele were confirmed by PCR using primers (P5 + P6) (Fig. 3 -5 B) and by DNA sequencing. Western blot analysis using anti Flag antibody revealed translation of a Flag-t agged protein (Fig. 3 5 C). The flag-immunoreactive band is detected between ~55 and 60 kDa, consistent with the calculated molecular mass of ZIP14 (54 kDa) plus the 3Flag (~ 3 kDa). Moreover, the Flag-immunoreactive band could be knocked down by using siR NA targeting ZIP14, thus confirming the band as ZIP14 (Fig. 3 -5 C). Subcellular Localization of ZIP14 in HepG2 C ells Previous immunofluorescence studies in primary mouse hepatocytes detected Zip14 at the plasma membrane of nonpermeabilized cells (134) Here I used immunofluorescence analysis of HepG2 cells expressing endogenous ZIP14-3Flag to further investigate the subcellular localization of ZIP14 (Fig. 3 6 ). In permeabilized HepG2 cells, I detected ZIP14 -3Flag in intracellular puncta, as well as on the plasma membrane (Fig. 3 6 B). The specificity of the anti Flag antibody was confirmed by the absence of immunofluorescent staining in permeabilized wild-type HepG2 cells (data not shown). To determine if intracellul ar ZIP14 localizes to endosomes, I used Alexafluor 488 labeled human holo -TF, which is endocytosed by the cells (Fig. 3 6 C). As shown in Fig. 3 6 D, ZIP14 and the labeled TF partially colocalize in the cytosol. A colocalization rate of 56% between ZIP14 and TF was calculated by using the colocalization algorithm provided with the Leica-LSM SP5 software (Fig. 3 6 E). The presence of ZIP14 in TF -positive endosomes is consistent with the hypothesis that ZIP14 plays a role in the uptake of iron from TF

PAGE 68

68 Knockdown of ZIP14 in HepG2 Cells Decreases TBI U ptake To determine if endogenous ZIP14 plays a role in TBI uptake in HepG2 cells, I suppressed ZIP14 by using siRNA and then measured the cellular assimilation of 59Fe from 59Fe -TF. Efficient knockdown of ZIP14 was c onfirmed by Western blotting (Fig. 3 7 A). ZIP14 knockdown did not affect the levels of TFR1 or TFR2, which are both expressed in HepG2 cells (168) I was unable to detect DMT1 protein in these cells (data not shown). Suppression of ZIP14 resulted in 45% less uptake of 59Fe from 59Fe TF compared to control cells (Fig. 3 7 B). As the lower uptake of 59Fe could possibly reflect diminished uptake of TF I compared the amount of biotin-labeled TF taken up by control cells and after ZIP14 knockdown. Fig. 7 C shows that ZIP14 knockdown did not affect the uptake of TF. Effect of Holo -TF on the Abundance and C ell surface E xpression of ZIP14 in HepG2 C ells To determine if holo -TF altered ZIP14 protein levels, I treated HepG2 cells overnight with either holo-TF or apo -TF, and then measured levels of total cellular ZIP14 as well as cell -surface ZIP14. Total cellular ZIP14 levels did not change with or without treatment with holo -TF or apo -TF (Fig. 3 -8 ). However, cell -surface ZIP14 levels were higher in cells treated with holo -TF or apo-TF compared to untreated cells. Cell surface ZIP14 protein levels did not differ between holo -TF and apoTF -treated cells. The abundance of TFR2 was higher in cells treated with holo-TF than in apo-TF treated and untreated cells, consistent with previous studies (119,120) Detection of the plasma membrane protein Na+, K+ ATPase and the cytosolic protein tubulin served as positive and negative controls, respectively, for the isolation of cell -surface proteins.

PAGE 69

69 Discussio n ZIP14 mediates NTBI uptake into cells. A physiologic role for ZIP14 in NTBI uptake is suggested by the observation that human ZIP14 is exp ressed most abundantly in the liver, heart, and pancreas (132) the tissues that preferentially accumulate iron in ironoverload disorders such as hereditary hemochromatosis In addition to rapidly clearing NTBI, th e liver readily takes up TBI (162) Studies in isolated primary mouse and rat hepatocytes suggest that NTBI and TBI uptake may share a common iron transporter (141,160,164) A s tudy in HepG2 cells suggested that ZIP14 may represent this common tr ansporter (136) Specifically, it has been found that HFE expression resulted in downregulation of ZIP14, which was as sociated with decreased uptake of not only NTBI, but also TBI. Up to now, DMT1 is the only known iron transporter which can function in endosomal iron release. However DMT1 knockout and its loss of function mutation animal model s indicate that in hepatocyt es and possibly other cell types, there are other transporters which can function in this process (74,75,86) I tested the hypothesis that ZIP14 plays a role in the uptake of iron from TF specifically in endosomal iron release Firstly, I tested if increased expression of Zip14 would affect TBI uptake. To do this, I use d human HEK 293T cells an easily transfectable cell line A novel finding form our study was that overexpress ing Zip14 resulted in a 25% increased i ron uptake from TF Since DMT1 is the only known transporter functioning in endosomal iron release, one would predict that overexpression of DMT1 would enhance the uptake of T F -bound iron. However, Wetli et al. (169) found no increase in TBI uptake in HEK 293T cells stably overexpressing DMT1 Shindo et al. (138) also found no change in TBI uptake in HLF cells (a hepatoma

PAGE 70

70 cell line) which stably overexpress DMT1. Another recent study by overexpressing DMT1 in CHO cells found no difference in TB I uptake between control and transfected cells (170) One of the explanations by the authors is that DMT1 may not be the limiting factor for TBI uptake; instead, other proteins such as TF R 1 might limit the uptake of TF bound iron. Here, I report that overexpression of mZip14 in HEK 293T cells can augment the TBI uptake. T he d ifference between the effects of Zip14 and DMT1 overexpression on TBI uptake in cells implies the unique function of Zip14 in TBI uptake. It was also showed by overexpressing mZip14 in HEK 293T cells that Zip14 can transport iron at acidic pH. Especially at pH 6.5, mZip4 transfected cells can take up almost double the amount of TBI compared with control cells. Acidification of endosomes is required for dissociation of iron from T F (171) A study has shown that at pH 6.5, about 50% of ir on atoms dissociate from TF and the rest is removed from T F at pH 6.0 (42) The ability to transport iron at acidic pH implies tha t Zip14 is able to function in endosomes. Zip14 is most abundantly expressed in the liver (132) To study the role of Zip14 in hepatocytes, I used the hepatoma cell line, HepG2 cells, which are very similar to hepa tocytes in many respects including secretion of hepatocyte specific proteins (172) As in human liver, ZIP14 is abundantly expressed in HepG2 cells Hepatocytes and hepatoma cells are known to take up holo -TF by several mechanisms including a high affinity, low -capacity TFR1 mediated endocytic pathway and a low affinity, high -capacity TFR1 -independent pathway (173 -175) In primary mouse hepatocytes and HuH7 hepatoma cells, iron uptake via the TFR1-med iated

PAGE 71

71 pathway saturates at low TF concentrations between 50100 nM (108,109) Here it has been document ed that in HepG2 cells incubated with 100 nM TF knockdown of endogen ous ZIP14 resulted i n 45% less uptake of iron from TF compared to controls Herbison et al. (176) showed that knockdown of TFR1 in HuH7 cells resulted in 80% less uptake of iron from 50 nM TF compared to controls. T he reduced uptake was due to diminished uptake of TF because TF uptake was also 80% lower. In our study, knockdown of ZIP14 in HepG2 cells did not affect the uptake of TF but it did result in a 45 % lower uptake of iron. These data provide strong support that ZIP14 plays a role in the uptake of iron from TF presumably through the high affinity, low capacity TFR1 mediated endocytic pathway. It was further shown that ZIP 14 in HepG2 cells localizes not only on the plasma membrane, but also to the endosome, where i t partially colocalized with internalized TF C ompetition between TBI and NTBI uptake by hepatocytes indicates that a common iron carrier participates in both pathways. A prere quisite for this is that the protein involved should have a plasma membrane and intracellular distribution. Subcellular localization study has shown that DMT1 in human hepatocytes located mostly in the cytoplasm but faintly in the cell membrane (138) In contrast, in this study Z IP 14 has been shown to distribute widely on the plasma membrane and intracell ularly. Moreover DMT1 is hardly detectable in hepatocytes. Taken together, these results strongly indicate that Z IP 14 may play an i mportant role in TBI uptake in h epatocytes. I also found that holo -TF had no effect on the abundance of ZIP14 in HepG2 cells From this observation it is predict ed that hepatic ZIP14 levels would not be altered in response to elevated plasma concentrations of holo -TF, such as in hereditary hemochromatosis.

PAGE 72

72 Until now, the only transmembrane transport protein implicated in TBI uptake by hepatocytes was DMT1 (177) Data from the present study suggest that ZIP14, which is expressed in at least 10 times greater amounts than is DMT1 in HepG2 cells, plays a role in TBI uptake by hepatocytes.

PAGE 73

73 F igure 31. Overexpression of Zip14 increases TBI uptake. A) HEK 293T cells transfected with pCMVSport2 (control) or pCMVSport6/ mouse Zip14 were incubated with 100 nM 59Fe -TF for 4 h. Cells were harvested and cell associated radioactivity was determine -counting. The results are representative of one of three independent experiments without significant variation between experiments. B) Lysates of cells incubated with 100 nM 59Fe -TF for 4 h were analyzed by Western blott ing for Zip14, TFR1 and DMT1. To indicate protein loading among lanes, blot was stripped and reprobed for actin. Data are representative of one of three experiments.

PAGE 74

74 Figure 32. pH dependence of Zip14 -mediated iron transport HEK 293T cells transfected with pCMVSport2 (control) or pCMVSport6/mouse Zip14 were incubated with 2 M 59Fe -ferric citrate for 1 h in uptake buffer at pH 7.5, 6.5, and 5.5. The amount of 59Fe taken up by cells is expressed as cpm per mg of protein. Data represent the mean SEM of three independent experiment s.

PAGE 75

75 Figure 33 Subcellular localization of Z ip 14-GFP in HepG2 cells. A) Staining of nuclei DAPI. B) ZIP14 -GFP is detected at the plasma membrane and in intracellular puncta C) Detection of internalized holo-TF. Cells were incubated for 30 min with T exas red-labeled human holo -TF prior to fixation. D) Merged image of panels B-C to visualize colocalization of Z ip -GFP 14 and TF. The arrows indicate clear coloc alization of Zip14-GFP and TF. All images were obtained by using a Leica TCS SP5 laser -scanning confocal microscope. Cells were not permeabilized. Figure 34. Comparison of ZIP14 and DMT1 mRNA levels in HepG2 and HEK 293T cells. Total RNA was isolated from untreated HepG2 and HEK 293T cells. Transcript copy numbers wer e determined by using qRT -PCR Data represent the mean SEM of three independent experiments of triplicate samples.

PAGE 76

76 Figure 3 5 Targeted knock in of 3 Flag into the ZIP14 locus of HepG2 cells. A) Diagrams of the wild-type (WT) ZIP14 allele, the Flag + NeoR allele after homolog ous recombination with the targeting vector, and the Flag allele after excising the neomycin cassette with Cre recombinase. B ) PCR of genomic DNA identifies clones with the Flag + NeoR allele and Flag allele. C) Western blot analysis of HepG2 cells express ing F lag-tagged ZIP14. Knockdown of endogenous ZIP14 in HepG2 cells using siRNA targeting ZIP14. To indicate protein loading among lanes, blot was stripped and reprobed for actin .

PAGE 77

77 Figure 36 Confocal microscopic analysis of the subcellular localization of ZIP14 in HepG2 cells. A) Staining of nuclei using DAPI. B) ZIP14 is detected at the plasma membran e and in intracellular puncta. Endogenous ZIP14 in HepG2ZIP14 3Flag cells was detected in permeabilized cells by using anti Flag antibody followed by rho damine -conjugated secondary antibody. C) Detection of internalized holo -TF. Cells were incubated for 30 min with Alexafluor 488 labeled human holo -TF prior to fixation and permeabilization. D) Merged image of panels A -C to visualize colocalization of ZIP14 and TF. E) Areas of colocalization (designated by white) as determined using the colocalization tool provided with the Leica -SP5 software. All images were obtained by using a Leica TCS SP5 laser -scanning confocal microscope.

PAGE 78

78 F igure 37 Knockdown of ZIP 14 decreases TBI uptake by HepG2 cells A) Knockdown of ZIP14 in HepG2 cells does not affect the expression of TFR1 or TFR2. HepG2 cells were transfected with negative control (NC) or ZIP14 siRNA for 72 h prior to Western blot analysis for ZIP14, TFR1, TFR 2, and actin as a lane loading control. B) Knockdown of ZIP14 decreases the uptake of TBI. HepG2 cells transfected with NC or ZIP14 siRNA for 72 h were incubated with 100 nM 59Fe -TF for 4 h. Cells were harvested and cell associated radioactivity was determ -counting. The amount of TBI taken up by cells is expressed as cpm per mg of protein. Data represent the mean SEM of three independent experiments. C) Knockdown of ZIP14 does not affect the uptake of TF. HepG2 cells transfected with NC or ZIP14 siRNA for 72 h were incubated with or without 100 nM biotinlabeled holo -TF. After 4 h, cell lysates were analyzed by Western blotting for ZIP14, TF, and ac tin as a lane loading control. The results shown are representative of one of three experiments with out significant variation between experiments.

PAGE 79

79 Figure 38 Effect of holo -TF on the abundance and cell -surface expression of ZIP14 in HepG2 cells. HepG2 cells were incubated overnight in serum -free medium supplemented with 10 M apo -TF or 10 M holo -TF Total cell lysates and cell -surface proteins were obtained and analyzed by Western blotting for ZIP14, TFR2, Na+, K+ ATPase, and tubulin. TFR2 was used a s positive control for treatment with holo-TF. Na+, K+ ATPase and tubulin were used as markers for pl asma membrane and cytosolic proteins, respectively. The results shown are representative of one of four experiments without significant variation between experiments.

PAGE 80

80 CHAPTER 4 STRUCTURAL ANALYSIS OF ZIP14 Introduction ZIP14, a newly identified iron trans porter, can mediate both TBI and NTBI into cells. Although we are starting to understand the transport function of Z IP 14, very little is known about its structure. Determining the structural features of proteins is an essential part of understanding the ba sis of their function. ZIPs transport metal ion substrates across cellular membranes into cytoplasm. Most ZIP proteins are predicted to have eight transmembrane (TM) regions with a similar membrane topology in which both the amino(N -) and carboxy (C -) t erminal ends of the mature peptides are located on the outside surface of the plasma membrane. ZIP proteins range from 309 to 476 amino acids in length and this difference is largely due to the length between TM domains III and IV (178) A cyto plasmic loop between TM domains III and IV is relatively longer and often contains a histidine-rich domain as a potential metal -binding motif ; other loops between TM regions are quite short (130,178) Features of th is eight TM domain topology model have been studied for human ZIP1 and ZIP2 (hZIP1 and hZIP2). By overexpressing hemagglutinin antigen (HA) or 3 HA tagged recombinant hZIP1 or hZIP2 in K562 cells, it has been shown that both the aminoterminus of hZIP1 and carboxy terminus of hZIP2 are extracellular (179,180) Overexpression in HEK 293 cells of mouse Zip4 and Zip5 (mZip4 and mZip5) whose carboxy termini tagged with HA, indicated that the carboxy termini of both pr oteins localized extracellularly (181) However, all these studied ZIPs share no more than 35% consensus positions with Zip14 (hZIP1, 24.2%; hZIP2, 22.2%; mZip4, 31.5% mZip5, 33.5%). In one study, by using hydropath y values, minimal charge density, and four programs hZIP8 which

PAGE 81

81 shares 69% similarity with hZIP14, was predicted to have seven transmembrane domains with an extracellular N -terminus and intracellular C -terminus (182). In the same study, hZIP14 was also predicted to have seven transmembrane domains. There are several other studies which report ed the predicted topology models for ZIP14 protein. Three of them predicted ZIP14 contains eight transmembrane domains, among which one predicted hZIP14 topology (132) two predicted mouse Zip14 (mZip14, which shares 83% identity with hZIP14) (134,183) One study also predicted that both amino and carboxy termini of mZip14 were located extracellularly and the histidine -rich region -containing loop localized extracellularly as well (129) None of these models however, ha s been experimentally investigated. R e combinant human ZIP14 is sensitive to PNGase F digestion, indicat ing that it is a n N linked glycoprotein (132) N linked glycosylation [Glycosylation at the asparagine (Asn or N ) residue] is an important modification of protein, which starts co -translationally with the transfer of the pre -synthesized oligosaccharide chain from a lipid precursor to an asparagine residue of the nascent protein in the sequence Asn-XSer/Thr (in some case, it can be Asn-X-Cys), where X i s any amino acid except proline (184,185) After the initial glycosylation, the processing of the carbohydrate side chain takes place primarily in endoplasmic reticulum (ER) and the Golgi apparatus by the sequential addition and removal of monosaccharide units (186) Glycosylation on asparagines of membrane protein requires that potential asparagine is in the ER lumen during glycoprotein maturation, and is extracellular after protein incorporation into the plasm a membrane (187,188) Addition of carbohydrate side chains gives proteins branched and mobile polar domains, which helps cells secrete proteins of greater complexity and with

PAGE 82

82 better efficiency (189) N linked glycosylation has been proposed to have different functions in various proteins, such as serving as a signal for intracellular sorting and cell -cell interaction, helping protein folding and trafficking, promoting resistance to proteases, preventing protein aggregat ion and maintaining the proper conformation along with standard function of the protein (190,191) Removal of consensus glycosylation sequences or inhibition of glycosylation often results in misfolding or aggregati on of proteins that fail to reach the functional states (189,192) The misfolded proteins are often linked covalently to each other by disulfide bonds and their ultimate fate is degradation (193) It is found that not all the po tential asparagines sites can be glycosylated. In fact, only one third of the potential Asn-XSer/Thr sites in proteins are actually glycosylated and the efficiency of glycosylation depends on properly oriented and accessible Asn-X -Ser / Thr sequ on (186) In this chapter, I investigated mZip14 s membrane topology by both epitope mapp ing and computer program prediction. I also examined its glycosylation status and effect of N -linked glycosylation on its trafficking and iron transport activity. I found mZ ip14 has an extracellular N -terminus, the histi dine-rich l oop and the C -terminus are both intracellular, suggesting there are seven transmembrane domains of this protein. Mouse Zip14 is glycosylated at asparagines 52, 75, 85 and 100, residues that are all in the extracellular N terminus N linked glycosylation does not affect plasma membrane localization, but it is required for iron transport activity of m Zip14

PAGE 83

83 Results Structure of the murine Zip14 Gene and S equence A lignment of its T wo Protein I soforms T he murine slc39a14 gene ( mZip14 ) consists of 10 exons spread over more than 40 kb on chromosome 14. It has two transcripts that encode two isoform proteins of both 489 amino acids due to alternative splicing of the equal length of exon 5A or 5B (Fig. 4 -1A ). Sequence alignment indicates that the two isoforms share 95.9% identity and 97.5% consensus positions (Fig. 4 -1B). The high degree of conservation in amino acids sequences suggests their common function. I studied mZip14a the isoform that has been show n to transport iron (129) Epitope Mapping of mZip 1 4 Analysis of the amino acid sequence of mZip14 by using different prediction programs results in different sequences being identified as transmembrane heli c es (Ta ble 4 1). However, most programs predict a long N -terminus, a long extramembrane loop which includes a histidine rich region and a relativ e ly short C terminal tail. Prediction programs can sometimes incorrectly assign the topology of proteins, especially when the protein contains a signal peptide (154) S even different programs predict the presence of a signal peptide in mZip14 (Table 4 2). To investiga te the topology of mZip14 protein experimentally, immunofluorescence studies were performed to examine the orientation of N -, C termini and histidine-rich loop. HEK 293T cells were transiently transfected with constructs encoding mZip14 with a 3 Flag epi tope at the N terminus, C -terminus and the loop containing the histidine -rich region. When immunofluorescence studies were performed to detect N -terminal-3Flag mZip14, a signal was clearly detectable at plasma membrane in the absence of

PAGE 84

84 detergent s aponin (Fig. 4 -2B) No intracellular staining of EEA1 was detected, indicating that cells were not permeabilized (Fig. 4 2A). In contrast only in permeabilized cells, Flag signals could be detect ed in histidine -rich region and C -terminus 3Flag mZip14 constructs transfected cells (Fig. 4 -2C and 2D). I also tried to detect C -terminal -GFP mZip14 by GFP antibody after transient transfection of HEK 293T cells. Similar to the C terminal Flag tag, I could detect a signal only under permeabilized conditions (Fig. 4 -3A t o 3C), further supporting that the C terminus localizes intracellularly The opposite orientations of the N and C -termini implies that mZip14 contains an odd number of transmembrane regions, which is likely to be seven according to computer prediction re sults (Supported by programs TMHMM, HMMTOP and MINNOU, Table 4 -1). Enzymatic Analysis of G lycosylation Status of Zip14 To analyze the glycosylation status of mZip14 I first used NetNGlyc 1.0 and NetOGlyc 3.1 to predict the existence of N -linked and O -li nked glycans in mZip14. Mouse Zip14 is predicted to be glycosylat ed at five potential asparagine sites w ith no O linked glycosylation. In order to assess the role of N linked glycosylation in mZip14, HEK 293T cells were transiently transfected with non-Fla g tagged mZip14 ( p CMV Sport 6 mZip14) or N -T -Flag mZip14 encoding vector s for 24 h, cell lysate s w ere incubated with or without PNGase F before Western analysis. The band pattern was different from the control sample (without PNGase F treatment) in that the upper band disappears with PNGase F treatment, indicating the higher molecular mass band is a glycosylated form of mZip14 and the lower band represents the de glycosylated form (Fig. 4 4 A and 4 B). I also performed the same analysis by using the cell lysate s collected from HepG2 cells that endogeno usly express Flag-tagged ZIP14. I found that

PAGE 85

85 endogenous ZIP14 is also glycosylated as the ZIP14 band shifted down with PNGase F treatment (Fig.4 4C). This demonstrates that Zip14 is an N linked glycoprotein. Identification of N -linked G lycosylation Sites in mZip14 To identify which predicted asparagines are indeed glycosylated in mZip14, I mutated the five potential sites by replacing asparagines ( N ) with aspartic acid ( D ) singly or in combination. The effect of these mutations were examined by transiently expressing wide-type (WT) and mutants in HEK 293T cells, followed by Western analysis of the collected cell lysates. I observed that in the HEK 293T overexpression system, WT -mZip14 always exhibited two bands arou nd 50 kDa Single replacement of asparagines at positions 52, 75, 85 and 100 all led to a decrease in the molecular mass of the upper band, suggesting that each site is glycosylated whereas mutation of asparagi nes 455 did not affect the band pattern compared to WT. After removal of N glycosylation sites in tandem, the upper band showed a stepwise decrease as well. With d eletion of all four asparagines sites, only the lower band was detected (Fig. 4 -5A). The same result was observed by using the C -T -F lag mZip14 wide-type and mutant constructs ( Fig. 4 5B). These results indicate that the first four predicted asparagines near the amino terminus of mZip14 are indeed glycosylated, whereas asparagine 455 is not. Schematic Membrane Topology M odel of mZip14 Based on the epitope mapping results the identified glycosylation sites, and multiple computer program predictions (Table 4 1), A membrane topology model of mZip14 protein was proposed (Fig. 4 -6 ). In this model, mZip14 has seven transmembrane regions with an ex tracellular aminoterminus and an i ntracellular carboxy terminus. The histidine rich region is also located intracellularly.

PAGE 86

86 N -linked Glycosylation does not Affect Plasma Membrane Localization of mZip14 To determine the role of N linked glycosylation in its trafficking to the plasma membrane, HEK 293T cells were transiently transfected with wild -type ( WT ) N -T -Flag mZip14 (WT mZip14) and its ASNs mutant (which has all four glycosylated asparagines mutated to aspartic acids). Total cell lysate s and cell -su rface proteins b iotinylat ed with a membrane impermeable reagent NHS -SS -biotin were analyzed by Western blotting. In total cell lysate, WT mZip14 exhibited two bands whereas the ASNs mutant resulted in only one band of the lower molecular mass. A s trong sig nal of the mutant could also be detected at the cell surface (Fig. 4 -7A). The subcellular localization of the WT mZip14 and ASNs mutant w ere examined by confocal microscop y using the antibody against the Flag epitope under nonpermeabilized conditions Both wide -type and ASNs mutant could be detected on the plasma membrane (Fig. 4 7B). Cells transfected with empty vector had no detectable immunofluorescence (Data not shown). It was concluded that N -linked glycosylation of mZip14 is not required for the effici ent transport of the protein to the cell surface, at least in the HEK 293 overexpression system. N -lin ked G lycosylation is Required for the Iron Transport A ctivity of mZip14 It was found that the iron transport activity decreased significantly in the ASNs mutant -transfected cells compared to that of WT mZip14 -t ransfected cells (Fig 4 8A). Western analysis indicated that the protein expression level of the mutant was similar to WT mZip14 (Fig 4 -8B ). Discussion ZIP proteins have been predicted to have eight -TM domains (130,178) However, this does not apply to all the ZIPs. Human ZIP14 and ZIP8 have been predicted to be different from other ZIPs in that only one transmembrane (TM) domain exists instead of

PAGE 87

87 two at the positions of TM IV and TM V in other ZIPs (182) resulting in a seven -TM domain model for ZIP14 and ZIP8 proteins. Mouse Zip14 shares 88% consensus positions with hZIP14, and in the present study, I found that the N t erminus of mZip14 localized extracellularly and that the histidine-rich region and the C terminus were located intracellularly. These data support the proposed seven -TM domain model for Zip14. This model is also supported by computer prediction programs T MHMM, HMMTOP and MINNOU. Cuthbertson et al (194) examine d thirteen different membrane topology prediction programs. They predicted 40 membrane proteins of known 3D structure s and found that no method was consistently the best. Programs TMHMM and HMMTOP predicted the number and position of TM segments well with accuracies over 80%. In the eight -TM domain model for ZIPs, TM IV and V among all the ZIP proteins frequently contain conserved histidine residues with polar or charged amino acids nearby T hese two transmembrane helices are predicted to form a cavity through which the substrate passes (178) I n yeast system, these cons erved residues are essential for proper function (195) However in two ZIP members, hZIP8 and hZIP14, a conserved serine residue in TM IV, which has been reported as a functionally indispensable residue, was replaced by an asparagine (182) This asparagine is well conserved between mZip14 and hZip14, suggesting that Zip14 may be functionally different from other ZIPs. It was also demonstrated that the hist idine -rich metal binding motif localizes intracellularly, suggesting it is not involved in extracellular recruitment or binding of metal substrates. Indeed, this cytoplasmic histidine -rich domain has been shown to be essential for ubiquitindependent degradation hZIP4 protein (196)

PAGE 88

88 Co/Post -translational modifications are important for protein function. And it is important t o understand how these modifications affect t he activities and functions of proteins. Glycosylation is by far t he most abundant co/post translational event. In humans more than half of the proteins have carbohydrate molecules attached (197,198) Glycoproteins are components of extracellular matrices and cellular surfaces wh ere the carbohydrate side chains are implicated in cell -cell and cell -matrix recognition events (199) N -linked glycosylation has been shown to affect protein trafficking and/or proper function. Some membrane protei ns, including Ca2+ receptor and the glucagonlike peptide 1 receptor h ave been shown to be dependent on N linked glycosylation for cell surface trafficking and functional activity (200,201) The l utropin receptor an d the norepinephrine transporter were reported to require N linked glycosylation for cell surface localization, but not for the high affinity ligand binding or substrate transport activity (202,203) N linked glycos ylation is also reported to be dispensable for protein cell surface localization in several cases. For example, N linked glycosylation is not essential for the trafficking to plasma membrane of breast cancer resistance protein (204) Without N -linked glycosylation, the organic solute transporter subunit was still trafficked properly to the plasma membrane and fully functional (205) Interestingly, a study of the human HERG potassium channel protein show ed that N linked glycosylat ion was not required for the cell surface expression of functional protein, whereas the mutation of a nonglycosylat ed asparagine caused a protein-trafficking defect leading to intracellular retention (206) Poten tial N linked glycosylation sites can be identified by the presence of the Asn X Ser/Thr (X can not be proline) sequon in peptide sequences. But not all such sequons

PAGE 89

89 are glycosylated. It has been shown that minimum distance for o ligosaccharyltransferase re quired for N -linked glycosylation active site is a bout 1214 amino acid residues above the membrane and is oriented nearly parallel to the membrane surface (207) In the present study, the gel electrophoresis data indicated that all four potential asparagine sites at amino terminus are linked with glycan consistent with the ecto orientation of the aminoterminus. I further showed that mutation of all four N -linked glycosylation sites does not block the plasma membrane trafficking of mZip14 protein, but removing the glycosylation of mZip14 has functional consequences. It has been demonstrated here that the iron transport activity of mZip 1 4 decreased significantly without N linked glycan. Data present ed here provide for the first time e xperimental support for the proposed seven-TM domain model of Zip14, and provide evidenc e that N linked glycosylation is not required for membrane localization but it is important for iron t ransport activity of Zip14.

PAGE 90

90 F igure 4 1. Schematic representation of the mZip14 gene and sequence alignment of its two protein isoforms A) The mZip14 gene has 10 exons and encodes two transcripts due to alternative splicing of equal length of exon 5A or 5B. B) Two transcripts of mZip14 gene encode two isoform proteins, named mZip14a and mZip14b. Sequence alignment indicate that two isoforms are 95.9% i dentical and 97.5 % conserved ( i dentical amino acids are highlighted in light grey, dark grey indicates the conservative amino acid exchange)

PAGE 91

91 Figure 42. Flag epitope mapping of mZip14 HEK 293T cells were transient ly transfected with pCMVSport2 (contr ol) or Flag-tagged constructs A) EEA1 can only be detected under permeabilized condition, indicating the antibody can pass through cell membrane only after saponin treatment. B) N -T -FlagmZip14 can be detected by Flag antibody under both permeabilized and nonpermeabilized conditions. C -D ) His -Flag-mZip14 and C -T -Flag -mZip14 can only be detected by Flag antibody under permeabilized condition. Data are representa t ive of three independent experiments.

PAGE 92

92 F igure 43. The c arboxy terminus of mZip14 is intracellu lar HEK 293T cells were transient ly transfected with mZip14-GFP vector (GFP at C -terminus of mZip14). A) Green fluorescence signal of Zip14-GFP can be seen under both permeabilized and nonpermeabilized conditions. B) Zip14 -GFP can be detected by GFP antibo dy only after saponin treatment. C) Merge picture of panels A -B. Data represent two independent experiments.

PAGE 93

93 F igure 4 4. Deglycosylation of Zip14. A) HEK 293T cells were transi ent ly transfected with PCMVSport6/ mZip14. B) HEK 293T cells were transient ly transfected with N -T -FlagmZip14. C) HepG2-ZIP14 -Flag cell lysates. Five microgram s of protein were incubated with or without PNGase F before Western analysis. Overexpression of mZip14 in HEK 293T cells results in two bands by Western blot The upper band disappears with PNGase F treatment, indicating the higher molecular mass band is N linked glycosylated form of mZip14, while the lower band represents the de glycosylated form. Endogenous ZIP14 is also glycosylated as the ZIP14 band detected by Flag antibody shifted down with PNGase F treatment.

PAGE 94

94 F igure 4 5. Identification of N -linked glycosylation sites in mZip14. A) Western blot analysis of cell lysates from HEK 293T cells transiently transfected with empty vector (pCMVSport2, Sport2) or N -T -Flag mZip14 expression vectors. Zip14 expression constructs encode either wildtype (WT) or N -glycosylation site/sites mutants ( N52D, N75D, N85D, N100D, N455D, N52/75D, N52/75/85D and N52/75/85/100D). Position 52, 75, 85, and 100 are identified to be glycosylated, but position 455 is not B) Western analysis by using C -T Flag mZip14 WT construct and mutants also indicates that the first four asparagines are linked with sugar and the last one is not. In both A and B, the samples were electrophoresed on 10% polyacrylamid e gel, transferred to nitrocellulose and probed with anti Flag antibody for Zip14.

PAGE 95

95 F igure 46. Schematic illustration of mZip14 membrane t opology Mouse Zip14 has 7 transmembrane segments with an extracellular N terminus and an intracellular C -termi nus. Histidine -rich region, also intracellular, is indicated by individual white circle s of histidines (H). Black circles of asparagine (N) indicate N linked glycosylation sites.

PAGE 96

96 F igure 47 N -linked glycosylation does not affect plasma membrane locali zation of mZip14. A) HEK 293T cells were transiently transfected with empty pCMVSport2 vector (Sport2), wide -type N -T -FlagmZip4 (WT mZip14) or its mutant with all four glycosylated asparagines replaced with aspartic acids (ASNs mutant). After 24 h, total cell lysates were harvested by using SDS lysis buffer, while c ell -surface proteins were label ed with NHS -SSbiotin. Samples were analysis by Western blotting for Zip14. After stripping, blots were reprobed for Lamp1 and Na+, K+ ATPase as markers for cytoso lic and plasma membrane proteins, respectively. B) After transfection, both WT and ASNs mutant mZip14 are detected at the plasma membrane in nonpermeabilized cells. Zip14 signals were detected by using anti -Flag antibody followed by r hodamine -conjugated se condary antibody (r ed). Nuclei were stained with Dapi (b lue ). Data represent three independent experiments.

PAGE 97

97 Figure 48. Function al analysis of mZip14 lacking N -linked glycosylation. A) The iron transport activity of wild type and mutant mZip14 was analy zed by measuring the NTBI uptake 48 h after transient expression in HEK 293T cells. After three times washing and incubation with SFM for 1 h, cells transfected with empty pCMVSport2 (Sport2), N -T -Flag mZip14 (WT mZip14) or N -T -Flag mZip14 asparagines mutant which has disruption in all four glycosylated asparagines (ASNs mutant) were incubated with 2 M 59Fe-ferric citrate for 2 h in uptake buffer at 37C. The amount of 59Fe taken up by cells is expressed as cpm per mg of protein. Data represent the mean SEM of three independent experiments. B) Cell lysates harvested 48 h after transient transfection were analyzed by immunoblot for Zip14 and tubulin as a lane loading control.

PAGE 98

98 Table 4 1. Bioinformatic Prediction of mZip14 Transmembrane (TM) R egions Tabl e 4 2. Prediction of mZip14 Signal Peptide Cleavage S ite Programs Existence of Signal Peptide Cleavage Position Cleavage Site SignalP 3.0 Y 28/29 A/S SignalP V2.0.b2 Y 28/29 A/S SPEPlip Y 28/29 A/S Phobius Y 28/29 A/S Predisi Y 24/25 P/Q Sig Pred Y 24/25 P/Q Y 22/23 T/A SOSUI Y 21/22 R/T MEMSAT SVM Y 16/17 L/F

PAGE 99

99 CHAPTER 5 CONCLUSIONS AND FUTURE DIRECTIONS Conclusions The central hypothesis of my dissertation was that ZIP14 plays a role in TBI uptake, functioning in endosomal iron release. In HEK 293T cells, overexpression of mouse Zip14 protein, TBI uptake increased about 25% without increasing TFR1 and DMT1 levels. Zip14 transfected cells were also shown to transport twice the amount of iron give n as ferric citrate at pH 6.5, the pH at which iron atoms dissociate from TF within endosomes These data suggest that Zip14 may play a role in endosomal iron release. In H epG2 cells expressing recombinant Zip14-GFP GFP signal was clearly detectable at the plasma membrane and exhibited ab undant punctuate intracellular staining which partially colocalized with endocytosed TF, a recycling endosomal marker. Detection of Zip14 -GFP in recycl ing endosomes further implicates this protein in the assimilation of TBI. HepG2 cells a hepatoma cell l ine, express at least 10 times more ZIP14 compared with DMT1 measured by mRNA copy number, implying ZI P14 plays an important role in h epatocytes. Through an AAVmediated knock-in approach, I generated a HepG2 cell line expressing a Flag -tagged ZIP14 allele, allowing specific and sensitive detection of endogenous ZIP14 in these cells. Subcellular localization by c onfocal microscopic analysis of these cells detected ZIP14 at the plasma membrane and in the holo TF -containing endosomes consistent with the hypothesis that ZIP14 plays a role in the uptake of iron from TF Knockdown of endogenous ZIP14 with siRNA resulted in a 45% reduction in TBI uptake by HepG2 cells. These results suggest that

PAGE 100

100 ZIP14 participates in the uptake of iron from TF thus identifying a potentially new role for ZIP14 in iron metabolism The second part of my project was to study the structure of ZIP14 protein. I used mZip14 to study its membrane topology and glycosylation effects. To investigate the membrane topology of m Zip14 a Flag e pitope was inserted into the N -terminus, C terminus, as well as the long extramembrane domain containing a histidine-rich metal binding motif. The tagged proteins were expressed in HEK 293T cells, and the accessibility of the F lag tags by antibody was dete rmined by immunofluorescence analysis of intact and permeabilized cells. Based on the experimental results together with bioinformatic predictions, It was conclude d that m Zip14 has seven transmembrane domains, with an extracellular N terminus an intracell ular C -terminus and a cytoplasmic large loop which contains histidine -rich metal binding motif Furthermore, glycosylation sites were identified by mutating each of the 5 potential N -linked glycosylation sites. The mutants were transient ly expressed in HEK 293T cells, followed by Western blotting. I found that mZip14 is glycosylated at asparagines 52, 75, 85 and 100, residues that are all in the extracellular amino terminus, confirming mZip14 is an N -linked glycoprotein. Lastly, t o examine the role of glyc osylation in plasma membrane trafficking and iron transport activity, mZip14 wide -type and asparagine mutant (which has no N -linked glycosylation sites) were transiently expressed in HEK 293T cells I found that N glycosylation of mZip14 is not required fo r cell -surface localization but it is required for iron transport activity.

PAGE 101

101 Future Directions This study was performed exclusively in cell culture model s Future studies using whole animal s will be required to define the in vivo role of ZIP14 in iron met abolism especially in the liver Utilization of knockout mouse models has been powerful for reveal ing the function of genes in vivo Generation of a global and tissue -specific Zip14 gene knockout mouse will be needed to demonstrate the role of Zip14 in ir on biology. Comparative genomics has shown that the mouse and human genomes share high degree of homology, suggesting the mouse also serves as a model for finding new therapeutic interventions for human diseases (208) Iron disorder s such as hemochromatosis, is one of the genetic diseases attracting more attention from researchers The C282Y mutation in H FE gene leads to the most comm on form of hereditary hemochromatosis (2 09) S ince the discovery of the H FE gene in 1996 (140) much eff or t has been devoted to identify its precise function. In HH patients, excess iron deposits in the liver, mainly in hepatocytes. Neither the import no r the export pathway for iron in hep atocytes is well understood (9) H fe gene knockout ( H fe/ -) mice displayed liver iron-loading phenotype, similar to human HH disease The only known iron importer DMT1 was considered to account for the iron deposition, but s urvival rate of DMT1/ mice was greater when inactivating H fe together. H fe/ -DMT1/ animal continued to deposit iron in the liver during growth (86) These data suggest that iron transporter other than DMT1 is involved in hepatic iron loading throughout the development of HH. It has been found overexpression of HFE led to decreased Z IP 14 stability and reduced both TBI a nd NTBI uptake s in HepG2 cells, suggesting the interaction of these two prote ins. F urther research is needed to test this hypothesis by using gene knockout

PAGE 102

102 animals. H fe/ -Zip14/ -, DMT1/ -Zip14/ or H fe/ -DMT1/ -Zip14/ double or triple knockout animal will ideally be the better animal models. The structure of a protein is always important for understanding its biological function. The experimentally derived topology model of Zip14 from this study suggests that the histidine -rich metal binding motif is intracellular. Whether or not this motif is involved in iron binding or i ntracellular iron level sensing remains to be investigated. Another ZIP protein signature motif HEXPHEXGD is also present in Zi p 14 with the first histidine replaced with glutamic acid resulting in the sequence EE F PHE L GD It has been shown that this motif is important for metal binding and the glutamic acid substitution does not affect its ability to transport zinc (132) The Zip14 topology model from the present study suggests that this motif is within the third extracellular loop presenting the possibility that this motif is important for the recruitment and binding of extracellular iron, thus important for the iron transport activity. However this hypothesis needs to be tested. The signal motif of Zip14 for endocytosis is another interesting aspect. If ZIP 14 localizes in endosomes, a specific internalization signal may present in its pr imary sequence. Endocytosis of cell surface protein is largely dependent on specific internalization motifs located within the proteins cytoplasmic domains. For example the t yrosine -containing motif with the consensus sequence of YXX where r epresents a hydrophobic residue, is present in LDL receptor ( as NPVY) and in TFR1 (as YTRF ) (210,211) both proteins are well known to internalize via a receptor mediated endocytosis manner. Sequence analysis indicates that Z ip14 has a conserved YSDI motif in the cytoplasmic histidine -rich loop, which could serve as a p otential

PAGE 103

103 internalization signal. To examine the role of this YSDI motif mutants of tyrosine or other amino acids in this motif need to be made.

PAGE 104

104 LIST OF REFE RENCES 1. Crichton, R. R., and Charloteaux Wauters, M. (1987) Eur. J. Biochem. 164, 485506 2. Hentze, M. W., Muckenthaler, M. U., and Andrews, N. C. (2004) Cell 117, 285297 3. Aruoma, O. I., Halliwell, B., Laughton, M. J., Quinlan, G. J., and Gutteridge, J. M. (1989) Biochem J 258, 617620 4. Andrews, N. C. (1999) N. Engl. J. Med. 341, 1986 -1995 5. Smith, R. S. (1965) Arch. Dis. Child. 40 343363 6. Andrews, N. C. (2000) Nat. Rev. Genet. 1 208217 7. Siah, C. W., Ombiga, J., Ad ams, L. A., Trinder, D., and Olynyk, J. K. (2006) Clin. Biochem. Rev. 27, 5 16 8. Munoz, M., Villar, I., and Garcia -Erce, J. A. (2009) World J. Gastroenterol. 15 46174626 9. Andrews, N. C. (2008) Blood 112, 219230 10. Sharp, P., and Srai, S. K. (2007) World J Gastroenterol 13 47164724 11. McKie, A. T., Barrow, D., Latunde -Dada, G. O., Rolfs, A., Sager, G., Mudaly, E., Mudaly, M., Richardson, C., Barlow, D., Bomford, A., Peters, T. J., Raja, K. B., Shirali, S., Hediger, M. A., Farzaneh, F., and Simpson, R. J. (2001) Science 291, 17551759 12. Gunshin, H., Starr, C. N., Direnzo, C., Fleming, M. D., Jin, J., Greer, E. L., Sellers, V. M., Galica, S. M., and Andrews, N. C. (2005) Blood 106, 2879 -2883 13. Frazer, D. M., Wilkins, S. J., Vulpe, C. D., and Anderson, G. J. (2005) Blood 106 4413; author reply 4414 14. Ohgami, R. S., Campagna, D. R., McDonald, A., and Fleming, M. D. (2006) Blood 108, 1388 -1394 15. Han, O., Failla, M. L., Hill, A. D., Morris, E. R., and Smith, J. C., Jr. (1995) J. Nutr. 125, 1291 -1299 16. Glahn, R. P., and Van Campen, D. R. (1997) J. Nutr. 127, 642647 17. Swain, J. H., Tabatabai, L. B., and Reddy, M. B. (2002) J. Nutr. 132, 245-251

PAGE 105

105 18. Gunshin, H., Mackenzie, B., Berger, U. V., Gunshin, Y., Romero, M. F., Boron, W. F., Nus sberger, S., Gollan, J. L., and Hediger, M. A. (1997) Nature 388 482 488 19. Tandy, S., Williams, M., Leggett, A., Lopez -Jimenez, M., Dedes, M., Ramesh, B., Srai, S. K., and Sharp, P. (2000) J. Biol. Chem. 275 10231029 20. Hallberg, L. (1981) Annu. Rev. Nutr. 1 123-147 21. Shayeghi, M., Latunde -Dada, G. O., Oakhill, J. S., Laftah, A. H., Takeuchi, K., Halliday, N., Khan, Y., Warley, A., McCann, F. E., Hider, R. C., Frazer, D. M., Anderson, G. J., Vulpe, C. D., Simpson, R. J., and McKie, A. T. (2005) Cell 122, 789-801 22. Laftah, A. H., Latunde -Dada, G. O., Fakih, S., Hider, R. C., Simpson, R. J., and McKie, A. T. (2009) Br. J. Nutr. 101 11501156 23. Qiu, A., Jansen, M., Sakaris, A., Min, S. H., Chattopadhyay, S., Tsai, E., Sandoval, C., Zhao, R., Akabas, M. H., and Goldman, I. D. (2006) Cell 127, 917928 24. Raffin, S. B., Woo, C. H., Roost, K. T., Price, D. C., and Schmid, R. (1974) J. Clin. Invest. 54 1344 1352 25. Sharp, P., and Srai, S. K. (2007) World J. Gastroenterol. 13, 4716 -4724 26. An drews, N. C., and Schmidt, P. J. (2007) Annu. Rev. Physiol. 69, 69-85 27. McKie, A. T., Marciani, P., Rolfs, A., Brennan, K., Wehr, K., Barrow, D., Miret, S., Bomford, A., Peters, T. J., Farzaneh, F., Hediger, M. A., Hentze, M. W., and Simpson, R. J. (2000) Mol. Cell 5 299309 28. Vulpe, C. D., Kuo, Y. M., Murphy, T. L., Cowley, L., Askwith, C., Libina, N., Gitschier, J., and Anderson, G. J. (1999) Nat. Genet. 21 195-199 29. McClung, J. P., and Karl, J. P. (2009) Nutr. Rev. 67 100104 30. Hentze, M. W., and Kuhn, L. C. (1996) Proc. Natl. Acad. Sci. U. S. A. 93 8175 8182 31. Frishman, D., and Hentze, M. W. (1996) Eur. J. Biochem. 239 197-200 32. Guo, B., Phillips, J. D., Yu, Y., and Leibold, E. A. (1995) J. Biol. Chem. 270, 21645-21651

PAGE 106

106 33. Walland er, M. L., Leibold, E. A., and Eisenstein, R. S. (2006) Biochim Biophys Acta 1763 668689 34. Park, C. H., Valore, E. V., Waring, A. J., and Ganz, T. (2001) J. Biol. Chem. 276, 78067810 35. Nemeth, E., and Ganz, T. (2006) Annu. Rev. Nutr. 26 323 -342 36. Cohen, L. A., Gutierrez, L., Weiss, A., LeichtmannBardoogo, Y., Zhang, D. L., Crooks, D., Sougrat, R., Morgenstern, A., Galy, B., Hentze, M. W., Lazaro, F. J., Rouault, T. A., and Meyron-Holtz, E. (2010) Blood 37. Torti, F. M., and Torti, S. V. (200 2) Blood 99 3505 3516 38. Munro, H. N., Aziz, N., Leibold, E. A., Murray, M., Rogers, J., Vass, J. K., and White, K. (1988) Ann. N. Y. Acad. Sci. 526, 113123 39. Harrison, P. M., and Arosio, P. (1996) Biochim. Biophys. Acta 1275, 161-203 40. Muckenthaler, M. U., Galy, B., and Hentze, M. W. (2008) Annu. Rev. Nutr. 28 197-213 41. Aisen, P., and Listowsky, I. (1980) Annu. Rev. Biochem. 49 357-393 42. Nunez, M. T., Gaete, V., Watkins, J. A., and Glass, J. (1990) J. Biol. Chem. 265, 66886692 43. Kuhn, L. C., McClelland, A., and Ruddle, F. H. (1984) Cell 37 95 103 44. Jing, S. Q., and Trowbridge, I. S. (1987) EMBO J. 6 327-331 45. Andrews, N. C. (2000) Annu. Rev. Genomics Hum. Genet. 1 75 -98 46. Yamashiro, D. J., Tycko, B., Fluss, S. R., and Maxfi eld, F. R. (1984) Cell 37 789800 47. Sipe, D. M., and Murphy, R. F. (1991) J. Biol. Chem. 266, 8002 8007 48. Sendamarai, A. K., Ohgami, R. S., Fleming, M. D., and Lawrence, C. M. (2008) Proc Natl Acad Sci U S A 105, 7410 -7415 49. Chan, L. N., and Gerh ardt, E. M. (1992) J. Biol. Chem. 267, 8254 -8259 50. Bernstein, S. E. (1987) J. Lab Clin. Med. 110 690 -705

PAGE 107

107 51. Trenor, C. C., 3rd, Campagna, D. R., Sellers, V. M., Andrews, N. C., and Fleming, M. D. (2000) Blood 96, 1113 -1118 52. Levy, J. E., Jin, O., Fujiwara, Y., Kuo, F., and Andrews, N. C. (1999) Nat. Genet. 21, 396399 53. Knutson, M., and Wessling -Resnick, M. (2003) Crit. Rev. Biochem. Mol. Biol. 38 6188 54. Ryter, S. W., and Tyrrell, R. M. (2000) Free Radic. Biol. Med. 28 289 -309 55. Donovan A., Lima, C. A., Pinkus, J. L., Pinkus, G. S., Zon, L. I., Robine, S., and Andrews, N. C. (2005) Cell Metab. 1 191 -200 56. Knutson, M. D., Vafa, M. R., Haile, D. J., and Wessling -Resnick, M. (2003) Blood 102, 41914197 57. Wilkins, M., Williams, P., a nd Cavill, I. (1977) Ann. Rheum. Dis. 36 474475 58. Tolosano, E., Fagoonee, S., Hirsch, E., Berger, F. G., Baumann, H., Silengo, L., and Altruda, F. (2002) Blood 100 42014208 59. Nielsen, M. J., and Moestrup, S. K. (2009) Blood 114, 764771 60. Kris tiansen, M., Graversen, J. H., Jacobsen, C., Sonne, O., Hoffman, H. J., Law, S. K., and Moestrup, S. K. (2001) Nature 409, 198201 61. Lim, S. K., Kim, H., bin Ali, A., Lim, Y. K., Wang, Y., Chong, S. M., Costantini, F., and Baumman, H. (1998) Blood 92, 1 870-1877 62. Hvidberg, V., Maniecki, M. B., Jacobsen, C., Hojrup, P., Moller, H. J., and Moestrup, S. K. (2005) Blood 106, 25722579 63. Herz, J., and Strickland, D. K. (2001) J. Clin. Invest. 108, 779784 64. Tolosano, E., Hirsch, E., Patrucco, E., Cam aschella, C., Navone, R., Silengo, L., and Altruda, F. (1999) Blood 94 3906 -3914 65. Anderson, G. J., and Frazer, D. M. (2005) Semin. Liver Dis. 25 420 432 66. Bacon, B. R., and Tavill, A. S. (1984) Semin. Liver Dis. 4 181-192 67. Van Wyk, C. P., Lin der -Horowitz, M., and Munro, H. N. (1971) J. Biol. Chem. 246, 10251031

PAGE 108

108 68. Zhang, A. S., and Enns, C. A. (2009) Hematology Am. Soc. Hematol. Educ. Program 207214 69. Gruenheid, S., Pinner, E., Desjardins, M., and Gros, P. (1997) J. Exp. Med. 185 717-730 70. Gruenheid, S., Cellier, M., Vidal, S., and Gros, P. (1995) Genomics 25 514-525 71. Hubert, N., and Hentze, M. W. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 1234512350 72. Lee, P. L., Gelbart, T., West, C., Halloran, C., and Beutler, E. (1998) Blood Cells Mol. Dis. 24 199215 73. Andrews, N. C. (1999) Int. J. Biochem. Cell Biol. 31 991994 74. Fleming, M. D., Romano, M. A., Su, M. A., Garrick, L. M., Garrick, M. D., and Andrews, N. C. (1998) Proc. Natl. Acad. Sci. U. S. A. 95 1148 1153 75. Fleming, M. D., Trenor, C. C., 3rd, Su, M. A., Foernzler, D., Beier, D. R., Dietrich, W. F., and Andrews, N. C. (1997) Nat. Genet. 16 383-386 76. Sladic -Simic, D., Zivkovic, N., Pavic, D., Marinkovic, D., Martinovic, J., and Martinovitch, P. N. (1966) G enetics 53 1079 -1089 77. Oates, P. S., and Morgan, E. H. (1996) Am. J. Physiol. Gastrointest. Liver Physiol. 270 G826-832 78. Zhang, L., Lee, T., Wang, Y., and Soong, T. W. (2000) Biochem. J. 349, 289297 79. Ludwiczek, S., Theurl, I., Muckenthaler, M U., Jakab, M., Mair, S. M., Theurl, M., Kiss, J., Paulmichl, M., Hentze, M. W., Ritter, M., and Weiss, G. (2007) Nat. Med. 13, 448454 80. Su, M. A., Trenor, C. C., Fleming, J. C., Fleming, M. D., and Andrews, N. C. (1998) Blood 92, 2157 -2163 81. Touret, N., Martin -Orozco, N., Paroutis, P., Furuya, W., Lam Yuk -Tseung, S., Forbes, J., Gros, P., and Grinstein, S. (2004) Blood 104, 15261533 82. Canonne -Hergaux, F., Fleming, M. D., Levy, J. E., Gauthier, S., Ralph, T., Picard, V., Andrews, N. C., and Gros P. (2000) Blood 96, 3964 -3970 83. Canonne -Hergaux, F., Gruenheid, S., Ponka, P., and Gros, P. (1999) Blood 93 44064417

PAGE 109

109 84. Muir, A., and Hopfer, U. (1985) Am. J. Physiol. Gastrointest. Liver Physiol. 248, G376-379 85. Griffiths, W. J., Kelly, A. L., Smith, S. J., and Cox, T. M. (2000) QJ Med. 93 575587 86. Gunshin, H., Fujiwara, Y., Custodio, A. O., Direnzo, C., Robine, S., and Andrews, N. C. (2005) J. Clin. Invest. 115, 1258 -1266 87. Edwards, J. A., Sullivan, A. L., and Hoke, J. E. (1980) Blood 5 5 645648 88. Bowen, B. J., and Morgan, E. H. (1987) Blood 70 3844 89. Garrick, M. D., Gniecko, K., Liu, Y., Cohan, D. S., and Garrick, L. M. (1993) J. Biol. Chem. 268, 1486714874 90. Garrick, L. M., Dolan, K. G., Romano, M. A., and Garrick, M. D. ( 1999) J. Cell Physiol. 178 349-358 91. Mims, M. P., Guan, Y., Pospisilova, D., Priwitzerova, M., Indrak, K., Ponka, P., Divoky, V., and Prchal, J. T. (2005) Blood 105, 1337 -1342 92. Lam Yuk -Tseung, S., Mathieu, M., and Gros, P. (2005) Blood Cells Mol. D is. 35 212-216 93. Iolascon, A., d'Apolito, M., Servedio, V., Cimmino, F., Piga, A., and Camaschella, C. (2006) Blood 107 349 -354 94. Beaumont, C., Delaunay, J., Hetet, G., Grandchamp, B., de Montalembert, M., and Tchernia, G. (2006) Blood 107, 4168 -4170 95. Canonne -Hergaux, F., Zhang, A. S., Ponka, P., and Gros, P. (2001) Blood 98 38233830 96. Picard, V., Govoni, G., Jabado, N., and Gros, P. (2000) J. Biol. Chem. 275, 35738-35745 97. Gruenheid, S., Canonne-Hergaux, F., Gauthier, S., Hackam, D. J., Grinstein, S., and Gros, P. (1999) J. Exp. Med. 189, 831841 98. Tabuchi, M., Tanaka, N., Nishida Kitayama, J., Ohno, H., and Kishi, F. (2002) Mol. Biol. Cell 13, 4371 -4387 99. Tabuchi, M., Yoshimori, T., Yamaguchi, K., Yoshida, T., and Kishi, F. (2000) J. Biol. Chem. 275, 2222022228

PAGE 110

110 100. Richardson, D. R., and Ponka, P. (1997) Biochim Biophys Acta 1331 1 -40 101. Chua, A. C., Graham, R. M., Trinder, D., and Olynyk, J. K. (2007) Crit. Rev. Clin. Lab. Sci. 44 413459 102. Oates, P. S., Thomas, C., Fr eitas, E., Callow, M. J., and Morgan, E. H. (2000) Am. J. Physiol. Gastrointest. Liver Physiol. 278, G930936 103. Dong, X. P., Cheng, X., Mills, E., Delling, M., Wang, F., Kurz, T., and Xu, H. (2008) Nature 455 992-996 104. Bargal, R., Avidan, N., BenAsher, E., Olender, Z., Zeigler, M., Frumkin, A., Raas -Rothschild, A., Glusman, G., Lancet, D., and Bach, G. (2000) Nat. Genet. 26, 118123 105. Chen, C. S., Bach, G., and Pagano, R. E. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 6373 -6378 106. Kohgo, Y., Ikuta, K., Ohtake, T., Torimoto, Y., and Kato, J. (2008) Int. J. Hematol. 88, 7 15 107. Selden, C., Khalil, M., and Hodgson, H. J. (1999) Gut 44 443446 108. Trinder, D., Zak, O., and Aisen, P. (1996) Hepatology 23 1512 -1520 109. Chua, A. C., Herbison, C. E., Drake, S. F., Graham, R. M., Olynyk, J. K., and Trinder, D. (2008) Hepatology 47, 1737 1744 110. Goldenberg, H., Seelos, C., Chatwani, S., Chegini, S., and Pumm, R. (1991) Biochim Biophys Acta 1067 145-152 111. Morgan, E. H. (1991) Comp Bioche m. Physiol. A Comp. Physiol. 99 9195 112. Eisenstein, R. S. (2000) Annu. Rev. Nutr. 20 627662 113. Herbison, C. E., Thorstensen, K. D., Chua, A. C., Graham, R. M., Leedman, P. J., Olynyk, J. K., and Trinder, D. P. (2009) Am. J. Physiol. Cell Physiol. 114. Sciot, R., Paterson, A. C., Van den Oord, J. J., and Desmet, V. J. (1987) Hepatology 7 831 -837 115. Kawabata, H., Yang, R., Hirama, T., Vuong, P. T., Kawano, S., Gombart, A. F., and Koeffler, H. P. (1999) J. Biol. Chem. 274 20826-20832

PAGE 111

111 116. Fle ming, R. E., Migas, M. C., Holden, C. C., Waheed, A., Britton, R. S., Tomatsu, S., Bacon, B. R., and Sly, W. S. (2000) Proc. Natl. Acad. Sci. U. S. A. 97 2214 2219 117. Kawabata, H., Germain, R. S., Vuong, P. T., Nakamaki, T., Said, J. W., and Koeffler, H. P. (2000) J. Biol. Chem. 275, 16618-16625 118. West, A. P., Jr., Bennett, M. J., Sellers, V. M., Andrews, N. C., Enns, C. A., and Bjorkman, P. J. (2000) J. Biol. Chem. 275, 3813538138 119. Robb, A., and Wessling-Resnick, M. (2004) Blood 104 4294 4299 120. Johnson, M. B., and Enns, C. A. (2004) Blood 104, 4287 4293 121. Chua, A. C., Delima, R. D., Morgan, E. H., Herbison, C. E., Tirnitz -Parker, J. E., Graham, R. M., Fleming, R. E., Britton, R. S., Bacon, B. R., Olynyk, J. K., and Trinder, D. (2010) J. Hepatol. 52 425431 122. Wrighting, D. M., and Andrews, N. C. (2008) Curr. Top. Dev. Biol. 82 141167 123. Fleming, R. E., Britton, R. S., Waheed, A., Sly, W. S., and Bacon, B. R. (2005) Semin. Liver Dis. 25, 411419 124. Barisani, D., Berg, C. L., Wessling -Resnick, M., and Gollan, J. L. (1995) Am. J. Physiol. 269 G570-576 125. Hider, R. C. (2002) Eur. J. Clin. Invest. 32 Suppl 1 5054 126. Bernstein, S. E. (1987) J. Lab. Clin. Med. 110, 690-705 127. Hayashi, A., Wada, Y., Suzuki, T., and Shimi zu, A. (1993) Am. J. Hum. Genet. 53, 201213 128. Craven, C. M., Alexander, J., Eldridge, M., Kushner, J. P., Bernstein, S., and Kaplan, J. (1987) Proc. Natl. Acad. Sci. U. S. A. 84 3457 3461 129. Liuzzi, J. P., Aydemir, F., Nam, H., Knutson, M. D., and Cousins, R. J. (2006) Proc. Natl. Acad. Sci. U. S. A. 103, 1361213617 130. Guerinot, M. L. (2000) Biochimica et biophysica acta 1465 190198 131. Nomura, N., Nagase, T., Miyajima, N., Sazuka, T., Tanaka, A., Sato, S., Seki, N., Kawarabayasi, Y., Ishik awa, K., and Tabata, S. (1994) DNA Res. 1 223229 132. Taylor, K. M., Morgan, H. E., Johnson, A., and Nicholson, R. I. (2005) FEBS Lett. 579, 427-432

PAGE 112

112 133. Taylor, K. M., and Nicholson, R. I. (2003) Biochim. Biophys. Acta 1611 1630 134. Liuzzi, J. P., Lichten, L. A., Rivera, S., Blanchard, R. K., Aydemir, T. B., Knutson, M. D., Ganz, T., and Cousins, R. J. (2005) Proc. Natl. Acad. Sci. U. S. A. 102, 68436848 135. Grootveld, M., Bell, J. D., Halliwell, B., Aruoma, O. I., Bomford, A., and Sadler, P. J. (1989) J. Biol. Chem. 264, 4417 -4422 136. Gao, J., Zhao, N., Knutson, M. D., and Enns, C. A. (2008) J. Biol. Chem. 283 21462-21468 137. Garrick, M. D., Kuo, H. C., Vargas, F., Singleton, S., Zhao, L., Smith, J. J., Paradkar, P., Roth, J. A., and Garrick L. M. (2006) Biochem. J. 398 539-546 138. Shindo, M., Torimoto, Y., Saito, H., Motomura, W., Ikuta, K., Sato, K., Fujimoto, Y., and Kohgo, Y. (2006) Hepatol. Res. 35 152 -162 139. Olynyk, J. K., Cullen, D. J., Aquilia, S., Rossi, E., Summerville, L., and Powell, L. W. (1999) N. Engl. J. Med. 341 718-724 140. Feder, J. N., Gnirke, A., Thomas, W., Tsuchihashi, Z., Ruddy, D. A., Basava, A., Dormishian, F., Domingo, R., Jr., Ellis, M. C., Fullan, A., Hinton, L. M., Jones, N. L., Kimmel, B. E., Kronmal, G S., Lauer, P., Lee, V. K., Loeb, D. B., Mapa, F. A., McClelland, E., Meyer, N. C., Mintier, G. A., Moeller, N., Moore, T., Morikang, E., Prass, C. E., Quintana, L., Starnes, S. M., Schatzman, R. C., Brunke, K. J., Drayna, D. T., Risch, N. J., Bacon, B. R ., and Wolff, R. K. (1996) Nat. Genet. 13 399-408 141. Chua, A. C., Olynyk, J. K., Leedman, P. J., and Trinder, D. (2004) Blood 104 15191525 142. Potter, B. J., Blades, B., McHugh, T. A., Nunes, R. M., Beloqui, O., Slott, P. A., and Rand, J. H. (1989) Am. J. Physiol. 257, G524-531 143. Hirayama, M., Kohgo, Y., Kondo, H., Shintani, N., Fujikawa, K., Sasaki, K., Kato, J., and Niitsu, Y. (1993) Hepatology 18 874-880 144. Worthington, M. T., Browne, L., Battle, E. H., and Luo, R. Q. (2000) Am. J. Physio l. Gastrointest. Liver Physiol. 279, G1265 -1273 145. Zhang, X., Guo, C., Chen, Y., Shulha, H. P., Schnetz, M. P., LaFramboise, T., Bartels, C. F., Markowitz, S., Weng, Z., Scacheri, P. C., and Wang, Z. (2008) Nat. Methods 5 163165 146. Wang, Z. (2009) Methods Mol. Biol. 567, 87 -98

PAGE 113

113 147. Krogh, A., Larsson, B., von Heijne, G., and Sonnhammer, E. L. (2001) J. Mol. Biol. 305, 567-580 148. Tusnady, G. E., and Simon, I. (2001) Bioinformatics (Oxford, England) 17 849 850 149. Claros, M. G., and von Heijne, G. (1994) Comput. Appl. Biosci. 10 685686 150. Cao, B., Porollo, A., Adamczak, R., Jarrell, M., and Meller, J. (2006) Bioinformatics (Oxford, England) 22 303309 151. Hofmann, K., and Stoffel, W. (1993) Biol. Chem. Hoppe -Seyler 347 166 152. Arai, M. Mitsuke, H., Ikeda, M., Xia, J. X., Kikuchi, T., Satake, M., and Shimizu, T. (2004) Nucleic Acids Res. 32 W390-393 153. Jones, D. T., Taylor, W. R., and Thornton, J. M. (1994) Biochemistry 33, 3038 3049 154. Kall, L., Krogh, A., and Sonnhammer, E. L. (2007) Nucleic Acids Res. 35 W429432 155. Hirokawa, T., Boon -Chieng, S., and Mitaku, S. (1998) Bioinformatics (Oxford, England) 14 378379 156. Bendtsen, J. D., Nielsen, H., von Heijne, G., and Brunak, S. (2004) J. Mol. Biol. 340, 783-795 157. Farise lli, P., Finocchiaro, G., and Casadio, R. (2003) Bioinformatics (Oxford, England) 19 2498 -2499 158. Hiller, K., Grote, A., Scheer, M., Munch, R., and Jahn, D. (2004) Nucleic Acids Res. 32 W375-379 159. Hansen, J. E., Lund, O., Tolstrup, N., Gooley, A. A., Williams, K. L., and Brunak, S. (1998) Glycoconj. J. 15 115-130 160. Graham, R. M., Morgan, E. H., and Baker, E. (1998) Eur. J. Biochem. 253, 139145 161. Knutson, M. D. (2007) Nutr. Rev. 65 335-340 162. Morgan, E. H., Smith, G. D., and Peters, T. J. (1986) Biochem. J. 237 163173 163. Scheiber, B., and Goldenberg, H. (1996) Arch. Biochem. Biophys. 326, 185-192

PAGE 114

114 164. Trinder, D., and Morgan, E. (1997) Hepatology 26 691-698 165. McArdle, H. J., Danzeisen, R., Fosset, C., and Gambling, L. (2003) Biometals 16 161-167 166. Kralova, A., Svetlikova, M., Madar, J., Ulcova-Gallova, Z., Bukovsky, A., and Peknicova, J. (2008) Reprod. Biol. Endocrinol. 6 27 167. Trinder, D., Olynyk, J. K., Sly, W. S., and Morgan, E. H. (2002) Proc. Natl. Acad. Sci. U. S. A. 99 5622 -5626 168. Kawabata, H., Nakamaki, T., Ikonomi, P., Smith, R. D., Germain, R. S., and Koeffler, H. P. (2001) Blood 98 2714 -2719 169. Wetli, H. A., Buckett, P. D., and Wessling -Resnick, M. (2006) Chem. Biol. 13, 965-972 170. Zhang, A. S., Canonne -Hergaux, F., Gruenheid, S., Gros, P., and Ponka, P. (2008) Exp. Hematol. 36, 1227 1235 171. van Renswoude, J., Bridges, K. R., Harford, J. B., and Klausner, R. D. (1982) Proc. Natl. Acad. Sci. U. S. A. 79 6186 6190 172. Bokhari, M., Carnachan, R J., Cameron, N. R., and Przyborski, S. A. (2007) J. Anat. 211, 567-576 173. Ikuta, K., Zak, O., and Aisen, P. (2004) Int. J. Biochem. Cell Biol. 36 340-352 174. Morgan, E. H., and Baker, E. (1986) Fed. Proc. 45 2810 -2816 175. Trinder, D., Morgan, E. H., and Baker, E. (1988) Biochim Biophys Acta 943 440 446 176. Herbison, C. E., Thorstensen, K., Chua, A. C., Graham, R. M., Leedman, P., Olynyk, J. K., and Trinder, D. (2009) Am. J. Physiol. Cell Physiol. 297 C1567 1575 177. Gra ham, R. M., Chua, A. C., Herbison, C. E., Olynyk, J. K., and Trinder, D. (2007) World J. Gastroenterol. 13 47254736 178. Eide, D. J. (2004) Pflugers Arch. 447, 796800 179. Gaither, L. A., and Eide, D. J. (2000) J. Biol. Chem. 275, 55605564 180. Gait her, L. A., and Eide, D. J. (2001) J. Biol. Chem. 276, 22258-22264

PAGE 115

115 181. Wang, F., Kim, B. E., Petris, M. J., and Eide, D. J. (2004) J. Biol. Chem. 279, 51433-51441 182. Begum, N. A., Kobayashi, M., Moriwaki, Y., Matsumoto, M., Toyoshima, K., and Seya, T. (2002) Genomics 80 630 -645 183. Girijashanker, K., He, L., Soleimani, M., Reed, J. M., Li, H., Liu, Z., Wang, B., Dalton, T. P., and Nebert, D. W. (2008) Mol. Pharmacol. 73 1413 1423 184. Mellquist, J. L., Kasturi, L., Spitalnik, S. L., and Shakin -Eshl eman, S. H. (1998) Biochemistry 37, 6833 -6837 185. Miletich, J. P., and Broze, G. J., Jr. (1990) J. Biol. Chem. 265, 1139711404 186. Kornfeld, R., and Kornfeld, S. (1985) Annu. Rev. Biochem. 54 631664 187. Snider, M. D., and Robbins, P. W. (1982) J. Biol. Chem. 257 6796 6801 188. Kato, Y., Sakaguchi, M., Mori, Y., Saito, K., Nakamura, T., Bakker, E. P., Sato, Y., Goshima, S., and Uozumi, N. (2001) Proc. Natl. Acad. Sci. U. S. A. 98 6488 6493 189. Helenius, A., and Aebi, M. (2001) Science 291, 23642369 190. Nakamura, S., Takasaki, H., Kobayashi, K., and Kato, A. (1993) J. Biol. Chem. 268, 12706-12712 191. Helenius, A. (1994) Mol. Biol. Cell. 5 253 -265 192. Paulson, J. C. (1989) Trends Biochem. Sci. 14 272 -276 193. Helenius, A., Marquardt, T., and Braakman, I. (1992) Trends Cell Biol. 2 227231 194. Cuthbertson, J. M., Doyle, D. A., and Sansom, M. S. (2005) Protein Eng. Des. Sel. 18 295308 195. Rogers, E. E., Eide, D. J., and Guerinot, M. L. (2000) Proceedings of the National Academy of Sci ences of the United States of America 97 12356 -12360 196. Mao, X., Kim, B. E., Wang, F., Eide, D. J., and Petris, M. J. (2007) J. Biol. Chem. 282, 69927000 197. Apweiler, R., Hermjakob, H., and Sharon, N. (1999) Biochimica et biophysica acta 1473 4 -8 198. Ben -Dor, S., Esterman, N., Rubin, E., and Sharon, N. (2004) Glycobiology 14 95101

PAGE 116

116 199. Rudd, P. M., Elliott, T., Cresswell, P., Wilson, I. A., and Dwek, R. A. (2001) Science (New York, N.Y ) 291 2370 2376 200. Fan, G., Goldsmith, P. K., Collins, R ., Dunn, C. K., Krapcho, K. J., Rogers, K. V., and Spiegel, A. M. (1997) Endocrinology 138 1916 1922 201. Chen, Q., Miller, L. J., and Dong, M. (2010) Am. J. Physiol. Endocrinol. Metab. 299, E62 E68 202. Nguyen, T. T., and Amara, S. G. (1996) J. Neurochem. 67 645655 203. Liu, X., Davis, D., and Segaloff, D. L. (1993) J. Biol. Chem. 268, 1513-1516 204. Mohrmann, K., van Eijndhoven, M. A., Schinkel, A. H., and Schellens, J. H. (2005) Cancer Chemother. Pharmacol. 56 344-350 205. Soroka, C. J., Xu, S., Mennone, A., Lam, P., and Boyer, J. L. (2008) BMC Cell Biol. 9 57 206. Gong, Q., Anderson, C. L., January, C. T., and Zhou, Z. (2002) Am. J. Physiol. Heart Circ. Physiol. 283, H77 -84 207. Nilsson, I. M., and von Heijne, G. (1993) J. Biol. Chem. 268 57 98 5801 208. Guan, C., Ye, C., Yang, X., and Gao, J. (2010) Genesis 48, 73 -85 209. Brunt, E. M. (2005) Semin. Liver Dis. 25 392401 210. Girones, N., Alverez, E., Seth, A., Lin, I. M., Latour, D. A., and Davis, R. J. (1991) J. Biol. Chem. 266 19006 -19 012 211. Chen, W. J., Goldstein, J. L., and Brown, M. S. (1990) J. Biol. Chem. 265, 31163123

PAGE 117

117 BIOGRAPHICAL SKETCH Ningning Zhao was born in Jinhua, Zhejiang province, China. He received a Bachelor of Science degree and a Master of Education degree from Beijing Sport University, Beijing, China. He majored in e xercise b iochemistry and focused on sports n utrition He did a thesis research about iron and exercise induced anemia, which interested him to pursue further knowledge in iron metabolism. In June 2 006, he came to the Department of Food Science and Human Nutrition at University of Florida for the PhD program in nutritional s ciences. He worked with Dr. Mitchell Knutson and focused on molecular aspects of iron metabolism which broaden ed his u nderstanding about iron biology and inspired him to continue learning in the iron field. After graduation, he will go to Oregon Health and Science University for a postdoctoral training in iron biology at the laboratory of Dr. Caroline Enns.